## FROM BANKING TO SOVEREIGN STRESS — EXECUTIVE SUMMARY

## Source details

**Canonical URL:** [FROM BANKING TO SOVEREIGN STRESS — EXECUTIVE SUMMARY](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_122214.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_122214.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_122214.pdf.json)

---

### Purpose, scope, and key quantitative facts
- Purpose:
  - Explores how banking sector developments and characteristics influence propagation of risks from the banking sector to sovereign debt and the extent of fiscal costs of banking crises.
  - Proposes practices and policies for fiscal authorities to manage banking-related fiscal risks and enhance crisis preparedness.
  - Focus: emerging markets and advanced economies; relevance to some low income countries undergoing financial deepening.
  - Exclusions: nonbank financial sector vulnerabilities and migration of risks to the shadow banking system are beyond the scope.
- Key quantitative facts cited:
  - Laeven and Valencia (2013) database identifies 147 banking crisis episodes during 1970–2011.
  - Median cost of direct government intervention in the banking sector across the sample (1970–2011): about 7 percent of GDP.
  - Median increase in public debt when factoring indirect fiscal impact: about 12 percentage points of GDP (measured over the 4-year period from the precrisis year).
  - In almost a third of the sample, debt increased by more than 20 percentage points of GDP.

### Channels from banking stress to sovereign stress
- Principal channels:
  - Growth channel: rapid bank credit expansions and vulnerabilities amplify depth and length of recessions.
  - Fiscal balance channel: amplified recessions and weaker growth deteriorate primary balances.
  - Direct channel: fiscal cost associated with resolving crises (recapitalizations, asset purchases, guarantees).
  - Indirect channels: adverse impact on growth, interest rates, and asset prices that worsen fiscal outcomes.
- Mechanisms:
  - Deleveraging and private sector adjustment required after rapid banking expansions.
  - Creditless recoveries: recoveries preceded by credit booms and/or banking crises are weaker and more protracted.

### Determinants of fiscal costs and sovereign exposure
- Banking-sector characteristics associated with larger fiscal costs and deeper/worse recessions:
  - Larger banking sector size (bank assets-to-GDP).
  - Higher leverage (credit-to-GDP).
  - Greater reliance on wholesale or foreign funding (loan-to-deposit ratios; offshore deposits; loans from nonresident banks to GDP).
  - Higher dependence on foreign funding versus domestic deposits.
- Institutional and regulatory features:
  - Stronger bank supervision and broader deposit insurance coverage associated with lower direct fiscal costs.
  - Government guarantees on banks’ liabilities after crisis onset associated with higher direct fiscal costs and increased contingent liabilities.
  - Effective resolution frameworks and speedy recognition of loan losses with immediate recapitalization can lower direct and overall fiscal costs.
- Banks’ home bias (share of domestic sovereign claims in banks’ total assets):
  - Amplifies two-way feedbacks between banks and the sovereign.
  - Can reduce borrowing costs short term but may lead to increased accumulation of debt and eventual rise in sovereign risk.
  - Reflects preferential regulatory treatment of government debt, preference for local-currency liquid assets, and occasional government moral suasion.
- Market pricing and perceptions:
  - Markets reflect higher implied probability of sovereign distress conditional on bank stress in countries with more leveraged banking sectors.
  - Perception of government bailouts may contribute to adverse bank-sovereign links and higher sovereign spreads.

### Empirical findings — banking expansions, recessions, recoveries
- Sample and definitions:
  - Analysis based on 31 advanced economies and 29 emerging market economies, over 1970–2013.
  - “Severe” recession: output contraction amplitude and duration in top quartile of their distributions.
- Leverage and severe recessions:
  - In severe recessions, both private sector credit-to-GDP and loan-to-deposit ratios increased by some 25 percentage points.
  - Using an alternative severe recession definition, leverage ratios increased by 15 percentage points.
- Banking recessions (recessions preceded by significant 5-year banking expansion):
  - Pre-peak 3-year real GDP growth similar for banking and nonbanking recessions: about 11 percent.
  - Median banking recessions tend to be deeper and last longer than other recessions; emerging markets historically face deeper banking recessions than advanced economies.
  - Recoveries are slower after banking recessions: regaining pre-recession output takes about one more year versus nonbanking recessions.
- Severity of preceding banking expansions:
  - “Severe banking expansion” = top quartile of 5-year cumulative change in private sector credit-to-GDP within banking recession episodes.
  - Median recovery time after severe banking expansions is more than twice that after non-severe banking expansions.
- Role of external funding:
  - Rapid buildup of leverage often underpinned by increased reliance on wholesale funds.
  - Banking recessions associated with higher loan-to-deposit ratios and potentially greater reliance on external funding; such recessions tend to be deeper and longer.

### Fiscal outcomes of banking expansions and crises
- Fiscal balances and debt dynamics:
  - Rapid banking expansions associated with higher primary balances leading up to recession onset (particularly in advanced economies).
  - Once downturn begins, primary balances deteriorate more quickly in countries with prior rapid banking expansions.
  - In nonbanking recessions, median fiscal deterioration is reversed in four years; in banking recessions, five years after onset primary balances remain depressed and below pre-downturn levels.
  - Severe banking expansions reduce public debt during booms but subsequent fiscal deterioration during recessions more than offsets prior gains.
- Fiscal costs of systemic banking crises (1970–2011):
  - Median increase in public debt associated with banking crises: 12 percent of GDP.
  - Median direct fiscal costs associated with banking crises: 7 percentage points of GDP.
  - Costs vary considerably across countries.
  - Increase in public debt in advanced economies (sample median): almost twice the whole-sample median.
  - Sample of 25 banking crises starting 2007–11 (mainly advanced economies): median increase in public debt ≈ 18 percent of GDP.
  - Direct fiscal costs in advanced economies (average): about 4.2 percent of GDP.
    - In advanced economies, direct fiscal costs constituted 18 percent of overall debt increase.
    - For recent wave (mainly advanced economies), direct fiscal costs (about 4.2 percent of GDP) generally explained about 25 percent of total increase in public debt around the crisis.
- Components of fiscal costs:
  - Direct: bank recapitalizations; asset purchases; calls on government guarantees; depositor payouts; central bank recapitalization.
  - Indirect: revenue losses from lower growth and asset price declines; expenditure from automatic stabilizers; discretionary fiscal responses; market effects on borrowing costs; exchange rate effects.

### Determinants of direct and overall fiscal costs (regression evidence and robustness)
- Precrisis vulnerabilities associated with higher direct and overall fiscal costs:
  - Larger precrisis banking sectors (assets-to-GDP).
  - More leveraged banking sectors (credit-to-GDP).
  - Greater reliance on external funds and wholesale funding.
  - Higher loan-to-deposit ratios associated with higher overall costs.
- Institutional setting:
  - Direct fiscal costs lower in countries with better quality of supervision.
  - Direct fiscal costs lower where governments provide more generous financial safety nets (insured deposits over per capita GDP) — with moral hazard caveats.
- Resolution policies:
  - Guarantees on banks’ liabilities after crisis onset associated with higher direct and overall costs.
  - Imposing losses on shareholders and junior creditors can reduce need for taxpayer recapitalization.
  - Speedy loss recognition and immediate recapitalization/resolution can reduce fiscal costs.
- Robustness checks:
  - Results stable to including peak NPL ratio and depositor-loss dummies.
  - GMM instrumentation for policy responses (creditor rights variable; time dummies) preserves main findings: bank guarantees remain significant.
  - Timeframe variations (4, 5, 6 years) and different precrisis variable lags generally do not alter core results.

### Banks’ home bias: empirical evidence and implications
- Definition and data:
  - Home bias = banks’ holdings of domestic sovereign claims divided by banks’ total assets; sample: 22 AEs and 29 EMs; 1999–2012.
- Empirical findings:
  - Sovereigns in countries with high home bias generally face lower borrowing costs in normal times; negative relationship weakens or reverses when market sentiment (VIX) deteriorates.
  - Countries with home bias above the sample median have lower primary fiscal balances for a given level of lagged debt.
  - Countries with high home bias have higher debt levels and higher debt tolerance.
    - In 2007, average public debt level for countries in top quartile of home bias distribution was more than 50 percentage points of GDP higher than for those in lower quartile.
  - Panel regression coefficients (selected) from Box 2:
    - Bond spread, lagged: 0.129
    - Public debt/GDP, lagged: 0.134***
    - Output gap, lagged: 0.036
    - Inflation rate, 3-year MA, lagged: 0.542***
    - Institutional quality, lagged: -0.249***
    - Exchange rate depreciation, lagged: -0.007
    - Credit-to-GDP ratio, lagged: -0.030***
    - Credit-to-deposit ratio, lagged: -0.054
    - Capital account openness, lagged: -0.462
    - VIX: 0.037**
    - Constant: 28.56***
    - Observations: 250
    - Note: ***, **, * denote significance at 1%, 5%, and 10%.
- Interpretation:
  - Home bias can provide fiscal breathing space by reducing rollover risk and interest rates in normal times but may weaken fiscal responsiveness and encourage debt accumulation.
  - Under heightened global risk aversion, home bias can cease to lower—and may raise—borrowing costs.

### Market-based Bank-to-Sovereign Contagion (BSC) measure and determinants
- BSC construction (2005 Q1–2012 Q4; sample of 15 countries):
  - BSC = probability of sovereign default conditional on a median domestic banking default using CDS-implied marginal default probabilities and CIMDO joint distribution; conditional probabilities via Bayes’ law; median of P(A/B), P(A/C), ... used as dependent BSC.
- Key explanatory variables for BSC:
  - Growth forecast (expected to reduce contagion probability).
  - Short-term public debt/GDP (raises sovereign default risk).
  - Government bonds held by domestic banks/Assets (captures sovereign dependence on domestic bank financing).
  - Bank loans/GDP (scale); Banks capital/Assets (buffers); Price-to-book value (valuation).
  - Controls: EZProg dummy (post-2010 Q1), OMT dummy (on/after 2010 Q3), market price of risk (Mkt. Price of risk).
- Selected baseline estimation results (Panel FE, Column 1; 2005 Q1–2012 Q4):
  - Growth forecast: -0.112*** [-3.474]
  - Bank loans/GDP: 1.849*** [6.457]
  - Banks capital/Assets: -6.364* [-1.760]
  - Govt. bonds held by dom. banks/Assets: 8.847*** [3.853]
  - Gvt. Short-Term Debt/GDP: 2.833*** [7.853]
  - Mkt. Price of risk: 17.80*** [8.023]
  - EZProg: 0.334*** [8.429]
  - OMT dummy: -0.501*** [-4.052]
  - R-squared: 0.911
- Inference:
  - Higher market price of risk and higher bank exposures to government are associated with higher BSC.
  - Larger banking sector and greater short-term government debt associated with higher BSC.
  - Better bank valuation associated with lower BSC.
  - OMT announcements associated with lower BSC.
  - Growth forecasts and bank buffers mattered more pre-2010 Q1 than post-2010 Q1.

### Measurement tools and valuation of contingent liabilities
- Measurement approaches to contingent liabilities:
  - (i) face value of the guarantee;
  - (ii) expected cost (mean);
  - (iii) unexpected cost (maximum loss at specified confidence interval);
  - (iv) market value of the guarantee.
- Contingent Claims Analysis (CCA):
  - Risk-adjusted balance sheet framework using bank equity market information and balance sheet data to estimate default probabilities, expected losses, and guarantee-free credit spreads (FVCDS).
  - Difference between expected loss from FVCDS and observed bank CDS spread estimates market-implied government guarantee.
- Market- and rating-based methods:
  - Use observed market prices and model-based benchmarks; use ordinary and “stand-alone” credit ratings differences to infer implicit guarantees.
- Quick-estimate methodologies:
  - S&P’s Gross Problematic Assets metric for estimating contingent-liability risks.
- Applications:
  - CCA applied in GFSR April 2014, FSAPs, and IMF working papers.

### Good practices, institutional recommendations, and fiscal policy guidance
- Monitoring, measuring, reporting:
  - Establish mechanisms for information sharing between financial-sector monitors and fiscal authorities (while respecting confidentiality and regulator independence).
  - Build expertise within fiscal authorities to generate fiscal and stress scenarios related to banking vulnerabilities.
  - Use inter-agency analysis to inform policies enhancing resolution frameworks and reducing recourse to public support.
  - Routinely estimate impact of contingent liabilities on public debt whenever a new explicit guarantee is issued.
  - Design diagnostic tools to flag country-specific emerging bank-to-sovereign risks using triggers: leverage, size, holdings of domestic sovereign debt, funding mix, global risk aversion.
- Public discussion and disclosure:
  - Publish broad-based (not bank-specific) analysis informed by banking stress tests; present a range of scenarios including those that presume no use of public funds.
  - Benefits: increase awareness, build support for reforms, demonstrate action to reduce likelihood of taxpayer-funded bailouts.
  - Risks: misperception of guaranteed bailouts, potential to impede banks raising funds during stress; disclosure should be routine and carefully framed.
- Fiscal policy conduct:
  - Lean against the wind during banking booms to build fiscal buffers for countercyclical policies during downturns.
  - Adopt structural fiscal rules constraining spending of unsustainable revenue increases associated with credit booms (further work needed to develop such rules).
  - Balance benefits and risks of reliance on domestic banks for financing; diversify government financing sources during normal times.
  - Tax policy: consider limiting or eliminating tax deductibility of interest payments to reduce bias toward debt financing; IMF (2010) proposed a Financial Stability Contribution to internalize externalities and help build fiscal buffers.
- Regulatory and macroprudential policy:
  - Strengthen bank supervision, increase loss-absorbing capacity, implement effective resolution regimes, increase use of bail-in tools where appropriate.
  - Macroprudential instruments: credit-related (debt-to-income, loan-to-value caps); capital-related (countercyclical capital buffers, dynamic provisioning).
  - Institutional design for macroprudential authorities: access to information, clear mandate, cooperation while preserving autonomy.

### Survey findings on monitoring contingent liabilities and institutional practices
- Survey (sent Aug 7, 2013 to 117 market access countries; response rate 36.8% as of Oct 2013; 43 responses):
  - 77 percent of respondents have institutions monitoring explicit contingent liabilities; information reported to ministry of finance in almost all cases.
  - 60 percent indicated monitoring implicit contingent liabilities; follow-up showed only 12 percent do this regularly; 35 percent do it only under special circumstances.
  - 74 percent reported presence of crisis resolution frameworks; more than 60 percent of these frameworks explicitly foresee government intervention that could create sovereign liabilities.
  - In about 40 percent of countries with frameworks, practices for monitoring and quantifying implicit contingent liabilities are not in place; some of these countries have large and leveraged banking sectors, high public debt, or histories of banking crises.
  - Only 16 percent of respondents (7 countries) indicated cash buffers are estimated explicitly to cover potential financial sector risks.
- Policy-relevant observations:
  - Explicit contingent liabilities widely monitored; implicit contingent liabilities less systematically monitored.
  - DSAs and stress tests exist in many countries but seldom incorporate financial sector contingent liabilities.
  - Information sharing mechanisms exist in many jurisdictions but explicit links to fiscal and debt sustainability are often incomplete.

*Source: FROM BANKING TO SOVEREIGN STRESS — International Monetary Fund (executive summary and selected excerpts, December 22, 2014).*

### EXECUTIVE SUMMARY

### EXECUTIVE SUMMARY

### Purpose and scope
- Explores how banking sector developments and characteristics influence propagation of risks from the banking sector to sovereign debt and the extent of fiscal costs of banking crises.
- Proposes practices and policies for fiscal authorities to manage banking-related fiscal risks and enhance crisis preparedness.
- Focus: emerging markets and advanced economies, with relevance to some low income countries undergoing financial deepening.
- Exclusions: nonbank financial sector vulnerabilities and migration of risks to the shadow banking system are beyond the scope.

### Key quantitative facts cited
- Laeven and Valencia (2013) database identifies 147 banking crisis episodes during 1970–2011.
- Median cost of direct government intervention in the banking sector across the sample (1970–2011): about 7 percent of GDP.
- Median increase in public debt when factoring indirect fiscal impact: about 12 percentage points of GDP (measured over the 4-year period from the precrisis year).
- In almost a third of the sample, debt increased by more than 20 percentage points of GDP.

### Principal findings on channels from banking to sovereign stress
- Rapid bank credit expansions and associated banking sector vulnerabilities amplify depth and length of subsequent recessions (growth channel) and exert fiscal pressure via deterioration of primary balances (fiscal balance channel).
- These macro-financial and macro-fiscal channels operate during expansions and become more significant during banking crises.
- Banking crises transmit fiscal pressure through:
  - Direct channel: fiscal cost associated with resolving crises (recapitalizations, asset purchases, guarantees).
  - Indirect channels: adverse impact on growth, interest rates, and asset prices that worsen fiscal outcomes.

### Determinants of fiscal costs and sovereign exposure
- Banking sector characteristics associated with larger fiscal costs and deeper/worse recessions:
  - Larger banking sector size.
  - Higher leverage.
  - Greater reliance on wholesale or foreign funding.
- Institutional and regulatory features that mitigate or exacerbate fiscal costs:
  - Stronger bank supervision and broader deposit insurance coverage are associated with lower direct fiscal costs.
  - Government guarantees on banks’ liabilities after crisis onset are associated with higher direct fiscal costs and increase contingent fiscal liabilities.
  - Effective resolution frameworks and crisis management (speedy recognition of loan losses, immediate recapitalization) can lower direct and overall fiscal costs by enabling faster recoveries.
- Banks’ home bias (large holdings of domestic sovereign claims):
  - Amplifies two-way feedbacks between banks and the sovereign.
  - Can reduce borrowing costs and provide fiscal breathing space short term but may lead to increased accumulation of debt and eventual rise in sovereign risk.
  - Reflects factors including preferential regulatory treatment of government debt, preference for local-currency liquid assets, and occasional government moral suasion.
- Market pricing:
  - Markets appear to reflect higher implied probability of sovereign distress conditional on bank stress in countries with more leveraged banking sectors.
  - Perception that governments will bail out banks may contribute to adverse bank-sovereign links and higher sovereign spreads.

### Policy proposals and recommendations for fiscal authorities
- Strengthen financial sector regulation and supervision to preserve banking sector health and minimize taxpayer exposure; policy priorities include:
  - Macroprudential measures to reduce excessive procyclicality in banking systems.
  - Higher bank loss-absorbing capacities.
  - Effective resolution powers and planning.
- Institutional preparedness and monitoring:
  - Have in place an institutional framework that strengthens ability to identify and monitor banking-sector risks to public finances.
  - Design diagnostic analytical tools to assess emerging country-specific bank-to-sovereign risks (for example, early warning models or triggers for contingent liabilities analysis).
- Fiscal policy stance and rules:
  - Lean against the wind during banking booms to build fiscal buffers that allow appropriately-sized countercyclical policies during downturns.
  - Adopt fiscal rules that constrain spending of unsustainable increases in tax revenue associated with credit booms, including particular reliance on real estate-related sources of revenues; further work needed to develop such rules.
- Manage reliance on domestic bank financing:
  - Balance benefits and risks of relying on domestic banks as a source of public financing; excessive reliance can distort incentives, create a false sense of debt sustainability, and deepen bank-sovereign nexus.
- Tax policy to reduce incentives for leverage:
  - Consider tax policies that reduce bias toward debt financing and attractiveness of leverage (for example, removing tax incentives to borrow and introducing a Financial Stability Contribution tax) to lower banking sector risks and help build fiscal buffers when vulnerabilities increase.

*Source: EXECUTIVE SUMMARY (December 22, 2014) — FROM BANKING TO SOVEREIGN STRESS, International Monetary Fund*

### 9. Based on its findings, the paper highlights the importance of sound fiscal policies and

### 9. Based on its findings, the paper highlights the importance of sound fiscal policies and

### Importance of sound fiscal policies and monitoring fiscal risks from banking
- Paper highlights the importance of sound fiscal policies and practices to monitor and mitigate the fiscal risks stemming from the banking sector (Section IV).
- Survey of member countries suggests there may still be insufficient attention to fiscal risks created by the banking sector.
- Aim: raise awareness among policy makers and improve readiness to deal with potentially far-reaching fiscal implications of a buildup of ex-ante vulnerabilities in the banking sector, regardless of the strength of the regulatory framework.
- Proposes specific practices for fiscal authorities to monitor risks to public finances from the banking sector.
- Proposes that staff develop or build on existing analytical tools to help identify emerging bank-to-sovereign risks in the context of ongoing work to develop broader macro-financial surveillance toolkits.
- Reinforces calls for fiscal policies, including in the tax area, to mitigate banking booms and build up fiscal buffers in good times, thereby allowing for effective countercyclical policies during downturns and help safeguard fiscal sustainability.

### Role of regulatory, macroprudential, resolution, and fiscal policies
- Recognizes that banking regulatory and supervisory frameworks, including macroprudential and resolution policies, may be the best tools to safeguard banking sector health and reduce associated risks for public debt sustainability.
- Macroprudential and resolution policies are usefully complemented by appropriate fiscal policies that can reduce incentives to borrow and the scope for adverse feedback between sovereign risk and the financial system.
- Policy implications are generally relevant for most existing prudential and regulatory architectures because changes in regulatory frameworks typically lag financial market developments and risks can emerge from new sources.

### From banking developments to public debt stress — macro-financial and macro-fiscal channels
- Developments in the banking sector can impact debt sustainability by amplifying the depth and length of recessions, constraining the pace of recovery, and exerting strong fiscal pressures.
- Recent global crisis illustrates that buildups of excessive leverage, often associated with housing booms and fast credit growth, can lead to public debt sustainability concerns.
- Episodes of rapid credit growth can lead to excessive leverage that, when unraveled, amplify shocks on the real economy and the sovereign balance sheet.

### Banking sector expansions, recessions, and recoveries — key empirical findings
- Sample and methodology:
  - Analysis based on a sample of 31 advanced economies and 29 emerging market economies, over the period 1970–2013.
  - A recession is classified as “severe” when the output contraction from peak to trough (amplitude) and the duration (in quarters) from peak to recovery correspond to the top quartile of their distributions.
- Leverage and severe recessions:
  - In severe recessions, both private sector credit-to-GDP and loan-to-deposit ratios increased by some 25 percentage points.
  - Using an alternative severe recession definition, leverage ratios still increased considerably by 15 percentage points.
- Banking recessions (defined as recessions preceded by a significant banking expansion over 5 years):
  - During expansions, real GDP growth paths in the three years before output peak are similar for banking and nonbanking recessions, with output expanding in both cases by about 11 percent.
  - Median banking recessions tend to be deeper and last longer than other recessions; emerging markets have historically faced deeper banking recessions than advanced economies.
  - Recoveries are slower after banking recessions: regaining the pre-recession output level takes about one more year in banking recessions compared to nonbanking recessions for both advanced and emerging market economies.
- Severity of preceding banking expansions:
  - A “severe banking expansion” corresponds to the top quartile of the 5-year cumulative change in the private sector credit-to-GDP ratio within episodes of banking recessions.
  - Median recovery time for recessions preceded by severe banking expansions is more than twice that for recessions preceded by non-severe banking expansions.
  - The difference in amplitude (percentage change of real GDP between output peak and trough) further highlights high economic costs imposed by ex-ante banking sector vulnerabilities.
- Role of external funding:
  - Rapid buildup of leverage is often underpinned by increased reliance on wholesale funds.
  - Banks with loan-to-deposit ratios greater than 1 must fund loan portfolios through short-term wholesale funds, including external creditors that tend to be volatile.
  - Banking recessions are associated with higher loan-to-deposit ratios and potentially greater reliance on external funding; reliance on wholesale funding was a major source of bank vulnerabilities during the global financial crisis.
  - Banking recessions that followed expansions with increased reliance on foreign funding tend to be deeper and longer; large banking expansions often go hand in hand with large reliance on foreign funding.

### Mechanisms: deleveraging, creditless recoveries, and fiscal transmission
- Deleveraging and private sector adjustment:
  - Deeper and longer recessions after rapid banking expansions arise from banks’ need to quickly deleverage, necessitating sudden adjustment by the private sector.
  - Looser lending standards and asset price bubbles during expansions lead to excessive leverage and balance sheet weaknesses in the private sector, increasing economy sensitivity to shocks.
- Creditless recoveries:
  - Creditless recoveries tend to be preceded by credit booms and/or banking crises and those recoveries are weaker and more protracted than recoveries associated with credit expansion.

### Banking expansions and fiscal outcomes — empirical findings
- Fiscal balances during booms and recessions:
  - Rapid banking expansions are associated with higher primary balances leading up to the onset of the recession, particularly for advanced economies.
  - Once the downturn begins, primary balances deteriorate more quickly in countries that previously had rapid banking expansions.
  - In nonbanking recessions, it takes four years for the median fiscal deterioration (measured by deterioration of the primary balance) to be reversed.
  - In contrast, five years after the onset of banking recessions, primary balances continue to be depressed and stay significantly below levels achieved during the five years prior to the downturn.
- Public debt dynamics:
  - Severe banking expansions tend to reduce public debt during booms, but fiscal deterioration during the recession more than offsets all previous gains.
  - Severe banking expansions are associated with a faster reduction in public debt leading up to the economic downturn; once the recession hits, public debt increases are much more pronounced in countries that had severe banking expansions.
- Perception of boom-related revenues:
  - The relationship reflects the often mistaken perception that revenue increases associated with credit booms are structural.
  - If revenue increases are assumed structural and lead to increased expenditures that are difficult to unwind, loss of fiscal flexibility combined with automatic stabilizers and revenue losses during recessions reinforces negative impacts on fiscal balances and debt sustainability.

### Fiscal costs of systemic banking crises
- Banking expansions can turn into systemic banking crises and affect public finances through direct and indirect channels.
  - Direct channel: governments directly support the banking system with guarantees, recapitalizations, and other measures; these policies carry large costs and contribute to increasing public debt.
  - Indirect channel: crises raise risk premia, disrupt credit supply, reduce consumption and investment, and negatively affect aggregate demand, growth, and asset values; these effects reduce government revenues and create pressures on public spending.
- Comprehensive fiscal impact metric:
  - A comprehensive indicator of the impact of banking crises on public finances is the change in gross public debt.
  - Public debt captures direct budgetary costs, indirect costs through the real economy, and any cost recovery.
  - Changes in public debt are imprecise measures but help shed light on the role of precrisis risk factors and policy choices in explaining overall fiscal costs.

*From: FROM BANKING TO SOVEREIGN STRESS, INTERNATIONAL MONETARY FUND*

### 23. Systemic banking crises can carry significant fiscal costs. During 1970–2011, the overall

### 23. Systemic banking crises can carry significant fiscal costs. During 1970–2011, the overall

### Overview: magnitude and variation of fiscal costs
- Median increase in public debt associated with banking crises (1970–2011): 12 percent of GDP.
- Median direct fiscal costs associated with banking crises (1970–2011): 7 percentage points of GDP.
- Costs varied considerably across countries, indicating important country-specific factors.
- Increase in public debt in advanced economies (sample median): almost twice the whole-sample median.
- Sample of 25 banking crises that started over 2007–11 (mainly advanced economies): median increase in public debt ≈ 18 percent of GDP.
- Direct fiscal costs in advanced economies (average): about 4.2 percent of GDP.
  - In advanced economies, direct fiscal costs constituted 18 percent of the overall debt increase (with significant cross-country variation).
  - For the recent wave of crises (mainly advanced economies), direct fiscal costs (about 4.2 percent of GDP) generally explained about 25 percent of the total increase in public debt recorded around the crisis.

### Components and channels of fiscal costs
- Direct Fiscal Costs (examples listed):
  - Bank recapitalizations
  - Asset purchases
  - Calls on government guarantees
  - Depositor payouts
  - Central bank recapitalization
- Indirect Fiscal Costs (examples listed):
  - Revenue effect from lower growth and the decline in asset prices
  - Expenditure effect from automatic stabilizers
  - Discretionary fiscal policy (revenue and expenditure) in response to increasing economic slack
  - Market effects on borrowing costs
  - Effects through exchange rate changes

### Country and banking-system characteristics that affect fiscal costs
- Key characteristics affecting the fiscal impact of banking-to-sovereign links:
  - Precrisis vulnerabilities in the banking system
  - Pre-existing regulatory framework and institutional features
  - Type of resolution policies used in banking crises
  - Other features such as banks’ home bias (share of sovereign debt in banks’ total assets)
- Precrisis banking vulnerabilities and recession depth/duration:
  - Factors associated with deeper recessions and longer recoveries:
    - Increase in banking sector leverage (proxied by credit-to-GDP)
    - Reliance on wholesale funds (proxied by loan-to-deposit ratios)
    - Expansion of banking sector size (banking sector assets-to-GDP)
    - Higher dependence on foreign funding versus domestic deposits
    - Indicators of risk-taking (return on equity, non-interest income, net interest margin, insolvency measures)
  - Asset diversification is negatively associated with deeper recessions.
  - Degree of liability dollarization is positively associated with deeper recessions.
- Precrisis vulnerabilities and fiscal costs of crises:
  - Direct fiscal costs (percent of GDP) are higher in countries with:
    - Larger precrisis banking sectors (bank assets-to-GDP)
    - More leveraged banking sectors (precrisis credit-to-GDP)
    - Greater reliance on external funds (offshore deposits/domestic deposits)
    - Greater reliance on external wholesale funds (loans from nonresident banks to GDP)
  - Overall costs (change in public debt) are also higher in countries with large and leveraged precrisis banking sectors.
  - Results hold for the most recent wave of banking crises starting in 2007.

### Institutional setting and crisis management policies
- Precrisis institutional features:
  - Direct fiscal costs are lower in countries with better quality of supervision.
  - Direct fiscal costs are lower where governments provide more generous financial safety nets (proxied by ratio of insured deposits over per capita GDP).
  - Caveat: broader deposit insurance coverage may also be associated with higher likelihood of crises in some literature; net benefit of deposit insurance is ambiguous and moral hazard concerns matter.
- Resolution strategies and fiscal costs:
  - Providing guarantees on banks’ liabilities after crisis onset is associated with, on average, higher direct fiscal costs.
    - Possible reasons: reverse causality (severe crises prompt broader guarantees), guarantees may be not credible or sufficient, incentives for banks to take higher risks under guarantees, or guarantees called due to fiscal constraints.
    - Establishing deposit insurance before a crisis may be more effective than issuing new guarantees during a crisis.
  - Imposing losses on bank shareholders and junior debt holders (and creditors further up the hierarchy if manageable) can significantly reduce need for taxpayer recapitalization.
  - Speedy recognition of loan losses and immediate recapitalization or resolution of failing banks can lead to faster crisis resolution and ultimately lower direct fiscal costs.
    - Example: Japan’s banking crisis in the early 1990s — recapitalization waited until 1999 via government-sponsored preferred stocks; delay and loan “evergreening” likely increased fiscal cost.
  - Central bank actions matter:
    - Ability to respond quickly to liquidity needs can stem liquidity crunches and preempt need for recapitalization.
    - Providing liquidity to failing banks without resolution can postpone needed resolution and lead to moral hazard and higher eventual fiscal costs.
- Short-term measures with upfront costs (recapitalization, asset purchases) do not necessarily increase overall fiscal cost if they re-establish banking-sector functioning and enable asset recovery over time.
- Post-global financial crisis policy developments aiming to reduce public burden:
  - Increased loss absorbency
  - Enhanced supervision
  - Effective resolution regimes and planning
  - Increased use of the bail-in tool (e.g., EU Bank Recovery and Resolution Directive (BRRD) mandating creditor bail-in)
  - Some countries have constrained the state’s ability to assist banks (e.g., the United States and the European Union).

### Banks’ home bias in sovereign debt and fiscal outcomes
- Home bias: share of domestic sovereign debt in banks’ total assets has increased significantly since the global crisis, particularly in advanced economies with rising public debt.
- Drivers of home bias:
  - Preferential regulatory treatment of sovereign debt
  - Use of sovereign debt as collateral
  - Structural financial-sector characteristics, sovereign bond market liquidity, country savings rate, and government policies during stress
- Two-way causality between home bias and public debt:
  - Governments facing fiscal difficulties may induce banks to buy domestic sovereign debt.
  - High demand for domestic sovereign debt by banks can influence discretionary fiscal policy, lowering sovereign borrowing costs and expanding fiscal space.
- Empirical findings (AEs and EMs; panel regressions that account for endogeneity):
  - Sovereigns in countries with high home bias generally face lower borrowing costs.
    - The negative relationship weakens or might reverse when market sentiment (measured by the VIX) deteriorates; safe-haven countries may be an exception.
  - Countries with home bias above the sample median have lower primary fiscal balances for a given level of lagged debt.
    - Banks’ willingness to absorb sovereign claims and the lower interest rates reduce fiscal responsiveness to debt levels.
    - Reverse causality (moral suasion by sovereigns) may also contribute.
  - Countries with high home bias have higher debt levels and higher debt tolerance.
    - In 2007, the average public debt level for countries in the top quartile of home bias distribution was more than 50 percentage points of GDP higher than for those in the lower quartile.
    - High home-bias countries tend to experience debt difficulties only at very elevated debt levels.

*Source: Excerpt from the IMF document provided.*

### 32. Home bias provides fiscal breathing space, but risks creating incentives for countries

### 32. Home bias provides fiscal breathing space, but risks creating incentives for countries

### Home bias: benefits and risks
- Home bias reduces rollover risk and interest rates, providing fiscal breathing space by allowing governments flexibility to weigh the cost and benefit of a fiscal consolidation, especially during recessions.  
- Risks:
  - Strong domestic bank demand for sovereign debt may reduce government responsiveness to debt accumulation, potentially encouraging postponement of problems until debt reaches dangerously high levels.
  - With much public debt held domestically, debt stress could have severe and widespread repercussions on the local economy, particularly if effective domestic firewalls and crisis management mechanisms are not in place.
  - Excessive reliance on banks’ financing could crowd out domestic private credit, weakening economic activity and potentially increasing public debt.

### Empirical observations (figures and distributions)
- Distributional observations shown (2005–07):
  - Avg public debt for AEs and EMs (excluding Japan) is 76 percent of GDP in countries whose home bias falls in the top quartile of the distribution.
- Small set of numeric markers reproduced from figure context:
  - -0.5
  - -0.9
  - -2.2
  - -3
  - -2
  - -1
  - 0

### Market-based measures of bank-to-sovereign contagion (BSC)
- Markets reflect sovereign-bank linkages in investment choices and sovereign pricing; during stress, flight to quality benefited sovereign securities and introduced greater differentiation among sovereigns.
- Segoviano and Espinoza (2014) derive the probability of sovereign distress conditional on bank stress (BSC) using market-based probabilities of default for sovereigns and individual banks; BSC reflects markets’ perception of bank-originating risks priced into sovereign borrowing costs.
- There is a statistically significant relationship between variation in BSC and banking system characteristics and macroeconomic developments.

### Directional impacts on BSC (summary of empirical findings)
- Banking system characteristics:
  - Higher ratio of bank loans to GDP → higher potential contagion from banks to the sovereign.
  - Higher share of sovereign debt instruments on banks’ balance sheets → higher potential contagion.
  - Higher bank capital buffers (capital-to-asset ratio) → lower contagion.
  - Higher bank valuations (price-to-book value) → lower contagion.
- Macroeconomic and fiscal developments:
  - Projected GDP growth, initial fiscal balance position, and short-term public debt to GDP are statistically significant in explaining BSC.
  - The price of risk (proxying global risk aversion) is positively related to BSC.
- Table 2 note: Signs of coefficients reported were consistent across regressions for different country groupings; estimation period 2005 Q1 to 2012 Q4.

### Heterogeneity across time and country groups
- Pre-Euro area debt crisis: GDP growth projections and banks’ capital buffers were important determinants of euro area BSC.
- During the crisis:
  - Eurozone fiscal situation became more relevant.
  - For countries in distress, the share of government bonds held by banks was especially important; price-to-book values mattered less.
  - Countries combining sound fiscal fundamentals and safer banks (favorable price-to-book values) fared well.
  - Policy interventions to stem contagion (for example, the ECB’s public announcement of readiness to conduct outright monetary transactions) proved crucial for stressed economies.

### Good practices for monitoring, measuring, and reporting risks from the banking sector to public debt
- Motivation: Despite improvements since earlier crises, gaps remain in managing fiscal risks from banks; the recent global crisis exposed insufficient pre-crisis consideration of financial sector risks to public finances and, in many cases, insufficient fiscal space to absorb contingent liabilities.
- Survey findings:
  - Many countries have mechanisms (formal and informal, MOUs, inter-agency committees) for information sharing; ministries of finance are generally represented and often chair or co-chair these forums.
  - Significant room for improvement exists in how authorities assess fiscal risks from the banking sector and use relevant information.
  - Risks to public debt from the financial sector are not adequately monitored and assessed prior to materialization; assessment of implicit contingent liabilities is often not done or is done only under impending stress.
  - Debt sustainability analyses rarely include financial sector contingent liabilities stress tests; when contingent liabilities are monitored, estimates are seldom used to inform public DSA stress tests.

### Institutional framework recommendations
- Establish mechanisms for information sharing between financial-sector monitors and fiscal authorities, with due consideration to confidentiality and regulator independence.
- Build expertise within fiscal authorities to generate alternative fiscal and stress scenarios related to banking-sector vulnerabilities.
- Use inter-agency analysis to inform policies that enhance the resolution framework and reduce recourse to public support (for example, demand higher loss absorption capacity from banks and, where necessary, firm-specific structural measures).
- In countries with large or rapidly expanding banking sectors and limited fiscal headroom, adopt resolution regimes that reduce the likelihood public funds will be needed.
- For internationally connected banking sectors, strengthen cross-border cooperation and information sharing, including:
  - Strengthening national resolution powers;
  - Agreeing firm-specific cross-border operational resolution strategies;
  - Agreeing on definition, size, and location of loss-absorbing buffers across borders;
  - Enhancing ability of home and key host authorities of G-SIFIs to maintain crisis management groups.

### Identifying and measuring relevant risks
- Key indicators to monitor (may vary by country): leverage, size, reliance on external funds; indicators discussed include:
  - Credit-to-GDP ratio increases;
  - Loan-to-deposit ratio;
  - Assets-to-GDP ratio;
  - Dependence on foreign funding.
- These indicators are associated with increased fiscal risk via deeper recessions, protracted recoveries, and larger direct banking crisis costs.
- Good practice: compare indicator evolution against cross-country benchmarks and country-specific historical performance; if vulnerabilities detected, incorporate corresponding stress tests into debt sustainability analysis.
- Techniques to estimate contingent-liability risks:
  - Identify face value of guaranteed liabilities;
  - Categorize explicit contingent liabilities by rough probability of being called;
  - Estimate net present values of implicit guarantees using contingent-claims analysis or banks’ credit ratings;
  - Estimate expected cost and “unexpected cost” or “cost-at-risk” (the greatest cost the government can expect with a given probability).
- Estimates should account for features of the crisis resolution framework that may mitigate risks, including potential bail-in options.
- Establish clear procedures for when to conduct in-depth assessments of fiscal risks from contingent liabilities; periodic disclosure rather than ad hoc disclosure during stress is recommended.
- Standing practice: estimate the impact of contingent liabilities on public debt whenever a new explicit guarantee of a bank’s liability is issued.
- Analytical tool recommendation: design a diagnostic tool to flag country-specific emerging bank-to-sovereign risks as part of macro-financial toolkits, using triggers such as leverage, size, holdings of domestic sovereign debt, funding mix, and global risk aversion.

*FROM BANKING TO SOVEREIGN STRESS — INTERNATIONAL MONETARY FUND*

### 47. Governments have been typically reluctant to publicly discuss fiscal risks from the

### _122214 - 47. Governments have been typically reluctant to publicly discuss fiscal risks from the

### Public discussion of fiscal risks from the banking sector
- Governments have typically been reluctant to publicly discuss fiscal risks from the financial sectors due to concerns that doing so could worsen moral hazard and increase the government’s fiscal costs in the event of a crisis.
- A belief in many economies was that the probability of a financial crisis with significant fiscal costs was too small or too speculative to warrant public discussion.
- Counterarguments and considerations:
  - Public discussion could help build support for reforms to tackle the problem. As the IMF’s Manual on Fiscal Transparency (2007) states, “Fiscal transparency helps to highlight potential risks to the fiscal outlook, resulting in an earlier and smoother fiscal policy response to changing economic conditions, thereby reducing the incidence and severity of crises.”
  - Publication of aggregate information on fiscal risks from the banking sector may:
    - Increase awareness within government and among the public.
    - Help build support for the government’s fiscal policy (e.g., its debt target) and reforms that reduce fiscal risks, including resolution regimes in line with international best practice.
    - Demonstrate that the government is aware of and has taken measures to reduce the likelihood that public funds are needed in future bank failures.
    - Reduce the perception that some bank failure or government intervention may be imminent when financial sector contingent liabilities are discussed, if reported routinely as part of broader fiscal risk discussion.
  - Risks of disclosure:
    - Could create misperceptions that government support would be forthcoming to bail out banks, worsening moral hazard or unhelpfully limiting government intervention options.
    - During a period of stress, disclosure of previously nonpublic information about banks’ problems could impede banks’ efforts to raise funds or, at worst, could trigger a bank run.
  - Recommended characteristics of published analysis:
    - Broad-based (not bank-specific).
    - Informed by the published results of banking sector stress tests.
    - Present a range of alternative scenarios, including scenarios that presume no use of public funds to support banks.

### Current practices and examples
- Most governments do not discuss fiscal risks associated with supporting the banking sector, though a few are beginning to include some discussion in public reports.
- Examples of cautious inclusion in public documents:
  - Philippines’ 2012 Fiscal Risks Statement.
  - Ireland’s National Risk Assessment 2014.
  - Italy’s 2012 Economic and Financial Document.
  - U.S. Financial Stability Oversight Council’s 2014 annual report (reports on policies designed to protect public finances from financial crises).
  - 2012 European Commission’s Fiscal Sustainability Report included a detailed assessment of the potential impact of banking losses on public finances under different regulatory scenarios.

### Conclusion and policy implications — overview
- Developments and vulnerabilities in the banking system can cause severe difficulties for sovereigns; links from the banking sector to public debt are part of a complex nexus with numerous channels and feedback loops.
- Bank stress can jeopardize fiscal sustainability through both direct and indirect channels, but well-designed sovereign interventions may be necessary to prevent crises from spinning out of control and undermining fiscal sustainability.
- While analysis focuses on EMs and AEs, several policy conclusions are relevant for LICs, notably:
  - Importance of having in place macroprudential policies to safeguard financial stability.
  - Need to monitor fiscal risks arising from the banking sector.

### Regulatory and macroprudential policy recommendations
- Using regulatory policies is the preferred approach to:
  - Preserve banking sector health.
  - Improve banks' loss-absorbing capacity.
  - Prevent banking crises.
- Recent financial sector regulatory reforms should help:
  - Strengthen bank regulation and supervision.
  - Increase loss absorbency (as a going concern and in resolution).
  - Lead to more effective resolution regimes.
- Resolution mechanisms relying on bank contributions would help limit risks to the sovereign when adequate funds are built to support bank resolutions without undue recourse to public funds.
- Further reforms could include measures emerging from ongoing review of the regulatory framework that governs banks’ economic incentives to hold government debt.
- The 2014 Triennial Surveillance Review found that Fund surveillance has made important progress in assessing risks and aspects of financial sector stability; going forward, analysis of financial sector issues could be better linked to fiscal developments and macro performance.

### Role and design of macroprudential policy (paragraphs 52–53)
- Adoption of appropriate macroprudential policies supports building buffers that help maintain financial stability:
  - Macroprudential policies complement macroeconomic policies and micro-prudential supervision.
  - They can mitigate procyclical feedback loops between asset prices and credit and strengthen the financial system’s capacity to absorb negative shocks.
- Examples of instruments:
  - Credit-related instruments: caps on debt-to-income and loan-to-value to stem credit growth.
  - Capital-related instruments: adjustments in the capital requirement and the risk weights of assets, and dynamic provisioning requirements to provide countercyclical buffers.
- Well-coordinated macroprudential policies should account for externalities across financial sub-sectors.
- Essential elements of a strong institutional framework for macroprudential policy:
  - Ensure ability to act by providing access to relevant information and assigning an appropriate range and reach of instruments to the macroprudential authority.
  - Foster willingness to act and counter biases for inaction by giving a clear mandate.
  - Promote effective cooperation in risk assessments and mitigation while preserving policy function autonomy.
- Need for practices to ensure adequate early warnings ahead of banking stress, because structural factors underlying bank vulnerabilities cannot be adjusted overnight.

### Institutional arrangements and monitoring (recommended actions)
- The paper argues for:
  - Putting in place an institutional framework that strengthens the ability of fiscal authorities to identify and monitor risks from the banking sector, including ensuring timely access to relevant information and developing expertise to evaluate potential impacts on debt sustainability.
  - Establishing clear procedures to undertake in-depth analysis of risks from banking sector contingent liabilities; a diagnostic analytical tool could be designed to highlight country-specific potential emerging bank-to-sovereign risks based on the paper’s analytical results and approaches used in other existing tools.
  - Routinely reporting to the public the risks to the sovereign from banks as part of a broader discussion of fiscal risks, taking care not to exacerbate moral hazard. Public discussion could be informed by banking sector stress tests and the existing regulatory framework including measures to build up buffers within banks (particularly systemically important ones) to limit recourse to public funds in case of stress.
  - The estimated fiscal cost of banking stress should reflect both direct and indirect costs under different scenarios without creating expectations that a particular scenario would prevail or signaling a particular intervention strategy by the government.

### Fiscal policy conduct to contain banking sector risks (paragraph 54)
- Broad policy guidance for fiscal policy conduct:
  - Lean against the wind during banking booms to provide buffers that would allow for appropriately-sized countercyclical policies during downturns.
    - Adoption of structural fiscal rules that constrain the spending of unsustainable increases in tax revenue associated with credit and asset price booms would be beneficial.
    - More work is needed to develop structural fiscal targets and rules that take into account factors beyond those reflected in the output gap, for example, credit growth and property and stock price dynamics.
  - Balance benefits and risks associated with reliance on domestic banks as a source of financing.
    - Under some circumstances, particularly during recessions, high banks’ home bias in sovereign debt may provide relief when other financing sources may not be available.
    - However, home bias can distort incentives, deepen the banks/sovereign nexus, contribute to excessive accumulation of debt, and eventually raise sovereign risk.
    - Policymakers and debt managers should diversify government financing sources, especially during normal times.
  - Aim to raise revenue in ways that can help reduce banking sector risks and build fiscal buffers as bank vulnerabilities increase.
    - Limiting or eliminating the tax deductibility of interest payments on mortgages and on debt instruments more generally could remove the bias toward debt financing and attractiveness of leverage—a proven contributing factor to the cost of crises.
    - Reducing the tax advantage of debt would encourage corporations and banks to use less debt and more equity, supporting regulatory efforts to increase banks’ equity capital.
    - IMF (2010) proposed a Financial Stability Contribution to internalize externalities, contain banks’ leverage and the size of potential contingent liabilities for the government, and help build up fiscal buffers whenever there is an increase in banking sector vulnerabilities. For countries so wishing, the stability contribution could help finance resolution mechanisms.

### Summaries of selected boxes
- Box 1 — Policies to Resolve Financial System Stress and Restore Financial Intermediation:
  - Well-timed government interventions can restore balance sheets and incentives, mitigate negative shocks, and help restart productive investment, but trade-offs exist between restoring intermediation and keeping taxpayer costs down while avoiding misallocations and preserving incentives.
  - Four key dimensions affecting effectiveness and fiscal cost of policy responses:
    - Having a sound framework for assuring financial sector stability (precrisis sanctions, special resolution regimes, deposit insurance, emergency liquidity schemes).
    - A swift response to minimize real economy impacts; avoid regulatory capital forbearance that delays recognition of losses.
    - Measures to address debt overhang of firms and households, including debt restructuring programs.
    - Steps to limit costs and moral hazard: shareholders first to absorb losses; creditors contribute via restructuring; link government support to privately raised capital.
  - Recapitalizations are a key policy instrument and account for most direct fiscal costs in past banking crises; design matters to avoid bailing out unviable institutions or creating broad moral hazard (comparative lessons from U.S. and Japan recapitalization experiences).
- Box 2 — Banks’ Preference for Domestic Sovereign Claims:
  - Sovereign debt often enjoys preferential regulatory treatment, underpinning banks’ holdings of domestic sovereign debt relative to other assets, especially amplified during downturns.
  - During stress, domestic sovereign debt becomes more important as central bank collateral and for secured wholesale funding; banks absorb sovereign debt as other asset quality deteriorates and private investment opportunities decline.
  - Policy actions can encourage banks to hold more government debt in stress (liquidity extension, central bank purchases, moral suasion).
  - Cross-country variation in home bias explained by structural characteristics: excess liquidity, institutional framework, political stability, institutional quality, deposit base size, and limited investment options.
  - Regression results (fixed effects panel, sample of 22 advanced and 29 emerging economies during 1999-2012) indicate:
    - Bond spread, lagged: 0.129
    - Public debt/GDP, lagged: 0.134***
    - Output gap, lagged: 0.036
    - Inflation rate, 3-year MA, lagged: 0.542***
    - Institutional quality, lagged: -0.249***
    - Exchange rate depreciation, lagged: -0.007
    - Credit-to-GDP ratio, lagged: -0.030***
    - Credit-to-deposit ratio, lagged: -0.054
    - Capital account openness, lagged: -0.462
    - VIX 4/: 0.037**
    - Constant: 28.56***
    - Observations: 250
  - Note: ***, **, * show significance at 1%, 5%, and 10%. VIX is the implied volatility of S&P 500 index options and proxies for global risk aversion.

*International Monetary Fund — From Banking to Sovereign Stress (excerpts).*

### Box 3. Using Quantitative Frameworks to Value Contingent Liabilities

### Box 3. Using Quantitative Frameworks to Value Contingent Liabilities

### Measurement approaches to contingent liabilities
- Contingent liabilities can be quantified in different forms, including measurement of:
  - (i) the face value of the guarantee (or maximum loss under the guarantee);
  - (ii) the expected cost of the guarantee (i.e., the mean of the distribution);
  - (iii) the unexpected cost of the guarantee (the maximum loss at a specific confidence interval); and
  - (iv) the estimated market value of the guarantee.
- Measuring (i) is the least data intensive and is generally used for pricing an explicit guarantee.
- Measuring (ii) through (iv) require implicitly or explicitly an assessment of the probability of a guarantee being called and typically rely on market or historical data (for example, estimating the expected cost of a guarantee through loan loss history) or the use of option pricing models such as the contingent claims analysis (CCA).

### Contingent Claims Analysis (CCA)
- CCA is a risk-adjusted balance sheet framework where bank assets equal equity plus risky debt (risky debt is default-free value of debt minus the expected loss to creditors of default).
- References cited within the box for the theoretical basis include Merton (1974 and 1977), and Gray, Merton, Bodie (2008).
- CCA uses bank equity market information, together with balance sheet data, to estimate:
  - default probabilities and expected losses;
  - guarantee-free credit spread—fair value CDS spread or FVCDS (reported by Moody’s Analytics).
- The difference between:
  - the expected loss associated with the FVCDS, and
  - the expected loss from observed bank CDS spread
  provides an estimate of the market-implied government guarantee (i.e., contingent liability).
- CCA can be used for the estimation of government guarantees including implicit guarantees to the financial sector (see Cebatori, 2008 “Contingent Liabilities: Issues and Practice”).

### Market- and rating-based methods
- Market-implied approaches: use observed market prices (e.g., CDS spreads) and model-based “guarantee-free” benchmarks (e.g., FVCDS) to infer the value of implicit guarantees.
- Credit-rating information:
  - Credit-rating agencies provide both an ordinary credit rating and a “stand-alone” rating that abstracts from the possibility of government support.
  - The difference in the two credit ratings, combined with information on the relationship between borrowing costs and credit ratings, gives an estimate of the value of the government’s implicit guarantee.

### Other quick-estimate methodologies
- S&P’s Gross Problematic Assets metric (which includes include overdue loans, restructured assets, foreclosed real estate and other assets recovered in loan workouts, and nonperforming assets sold to special-purpose vehicles) estimates the cost associated to contingent liabilities risks stemming from the financial sector.

### Applications and empirical evidence
- There is strong evidence that implicit and explicit government support for banks reduces banks’ borrowing costs and CDS spreads; bank creditors are beneficiaries of implicit and explicit government guarantees, while equity holders are generally not.
- The CCA approach has been applied to analyze too-important-to-fail contingent liabilities in:
  - GFSR April 2014,
  - Financial Sector Assessment Programs (FSAPs) (for example, the U.K. and Swedish FSAPs),
  - several IMF working papers (for example, Gray and Jobst 2013; and Gray and others 2013 with applications to the EU).

_Source: Box 3. Using Quantitative Frameworks to Value Contingent Liabilities_

### 6.       Regression results hold under a number of robustness checks. Specifically, we run

### 6. Regression results hold under a number of robustness checks

### Robustness checks on regressions of direct fiscal costs
- Robustness checks accounted for:
  - Size of the crisis shock and bail-in option: model extended to include the peak in the non-performing loan (NPL) ratio during the crisis and a dummy indicating whether losses were inflicted to depositors. Results remain broadly unchanged, with only the size of the banking sector becoming less significant.
  - Potential endogeneity and reverse causality: the basic equation (1) was estimated using GMM with instruments for policy responses consisting of the creditor rights variable constructed by Djankov (2008) and time dummy variables (Honohan and Klingebiel, 2003). Results from the basic regressions continue to hold, alleviating somewhat the likelihood of reverse causality problems in the interpretation of the findings.

### Overall fiscal costs: measurement and approach
- Metric used: change in gross public debt-to-GDP ratios as a comprehensive (though imperfect) indicator of the overall impact of banking crises on public finances. Rationale:
  - Changes in public debt ratios cover direct fiscal costs and indirect costs via GDP growth (e.g., revenue and automatic stabilizers) and governments’ financing costs.
  - Changes in overall debt capture costs that materialize over time (e.g., calls on guarantees, losses on government banking sector assets) and governments’ ability to recover part of initial outlays.
  - Caveat: changes in public debt capture debt not necessarily attributable, directly or indirectly, to banking crises.
- Focus on gross public debt because data are more widely available than net debt or net worth, and gross debt better captures government’s short-term exposure to financial stress.
- Empirical strategy: reduced-form model broadly based on the standard debt-flow equation. Overall fiscal costs measured as the change in public debt ratio over five years (between precrisis level and value observed at the fourth year after beginning of the crisis).

### Regression findings on overall fiscal costs
- Precrisis banking sector characteristics associated with higher overall fiscal costs:
  - Larger and more leveraged banking sectors: measured by the asset to GDP ratio and the credit to the private sector to GDP ratio respectively.
  - Higher loans-to-deposits ratio (indicator of banking sector leverage and funding mix) is associated with higher overall costs in contrast to the analysis of direct fiscal costs.
  - Other funding-mix indicators (offshore deposits to domestic deposits; loans from non-resident banks to GDP) also noted as relevant in some specifications.
- Role of banking supervision and interaction with funding mix:
  - When controlling for the quality of banking supervision, indicators of banking sector funding mix appear less relevant for overall fiscal costs. Interpretation: effective supervision can mitigate the importance of banks’ funding mix for overall fiscal costs, while size of the sector and private sector debt overhang remain important drivers—particularly for direct costs.
- Policy responses and overall costs:
  - Bank guarantees are unequivocally associated with higher overall costs (Table AII5; regression results summarized).
  - Asset purchases and bank recapitalizations—policies that may have a higher upfront cost—are not significantly correlated to an increase in public debt in benchmark regressions. This suggests that not all high upfront cost crisis-management policies turn out to be costly over time (though endogeneity concerns could affect this result).

### Robustness checks for overall fiscal costs regressions
- Alternative timeframes:
  - Results remain unchanged when varying the time span to calculate the change in public debt to 4 and 6 years.
  - The shorter the time span, the smaller the coefficients, suggesting that some costs materialize only after a few years from the beginning of the crisis.
  - Altering the timeframe of precrisis variables (different lags or precrisis averages rather than one-year lagged values) does not alter the main results much.
  - Exception: the financial sector leverage variable loses significance when averages are taken instead of the one-year lag, probably indicating a strong build-up of leverage just before the triggering of the crisis.
- Size of the crisis shock and bail-in options:
  - Introducing the peak NPL ratio during the crisis and a depositor-loss dummy yields results broadly similar to benchmark regressions; neither depositor loss nor NPL variables appear statistically significant.
- Endogeneity and reverse causality:
  - Re-estimating the basic equation (2) by the generalized method of moments and using as instruments for policy responses the creditor rights variable constructed by Djankov (2008) and time dummy variables (Honohan and Klingebiel, 2003) yields results suggesting:
    - Bank guarantees remain significant.
    - In some specifications, asset purchases and bank recapitalization dummies gain significance.

*Source: Fund staff estimates (from the IMF chapter/annex text).*

### 1.      Throughout the annex, we focus on the home bias indicator that reflects the banks’

### Throughout the annex, we focus on the home bias indicator that reflects the banks’ holding of domestic sovereigns claims in banks’ total assets as in Acharya and Steffen (2013).

### Data and definition
- Sample coverage: 22 advanced economies (AEs) and 29 emerging markets (EMs).
- Time horizon: 1999–2012 (annual frequency).
- Home bias indicator: numerator = banks’ holdings of domestic sovereign claims; denominator = banks’ total assets (includes domestic sovereign claims, claims on nonresidents, central bank and other sectors) — all based on the IMF IFS.
- For six countries (Australia, Canada, New Zealand, Norway, Switzerland, and the U.K.) the numerator is complemented with domestic banks’ holding of domestic sovereign claims based on Arslanalp and Tsuda (2012).
- Alternative home bias measure used in robustness: holding of domestic sovereign debt as a share of holdings of all sovereign debt.

### Macroeconomic variables and data sources
- Dependent variables: Long-term bond yields (In percent), EMBI stripped spreads (In percent), Debt-to-GDP (In percent), Primary balance to GDP ratio (In percent). Sources include IMF WEO / IFS and Bloomberg.
- Explanatory variables include lagged debt-to-GDP (In percent), GDP growth rate (In percent), inflation rates (Three year lagged moving average CPI inflation), exchange rate depreciation (In percent), Institutional quality index (ICRG), Credit-to-GDP ratio (In percent), Capital account openness (Chinn and Ito (2006)), VIX (Chicago Board Options Exchange Market Volatility Index), U.S. long-term bond yields (In percent), Output gap, Government expenditure gap, Trade openness, Oil price (log of trend oil price for oil exporters).
- VIX sample mean used: 22.7 point (AEs); for EMs VIX sample mean: 18.5 point.

### Methodology — sovereign borrowing costs (panel analysis)
- Estimation approach: two-step generalized method of moments (GMM) with instruments (housing price and lagged credit-to-GDP ratio) to address potential endogeneity of home bias.
- Model specification (equation (1)) includes: lagged public debt, home bias (entered as deviation from its sample median), interactions of home bias with debt and with VIX, and vector of macro controls.
- For AEs, spreads measured against German bonds (European countries) or U.S. bonds (non-European countries). For EMs, sovereign bond yields used.

### Findings — Sovereign borrowing costs (AEs)
- General result: For AEs with moderate to high debt levels, borrowing costs generally decline as home bias increases in normal times.
- Empirical detail: Panel regression results (3rd column of Table AIII2) show bond spreads are:
  - negatively influenced by the interaction of home bias and debt;
  - positively affected by the interaction of home bias and VIX.
- Threshold example: The negative home-bias×debt effect dominates during normal times leading to a decrease in spreads with higher home bias when the debt level is above 50 percent of GDP.
- Marginal effect: The higher the debt level and the higher the spread, the greater the marginal benefit of home bias on sovereign borrowing cost.
- Crisis dynamics: During crisis periods (risk aversion rises), the negative impact of home bias on spreads diminishes and may turn positive.
  - Example: when VIX is 80 points above the sample mean and debt is 60 percent of GDP, the interaction of home bias and VIX could surpass the interaction of home bias and debt.
- Mechanisms suggested: lower expected default when domestic banks hold sizable domestic sovereign claims; during high risk aversion banks demand higher risk premia while continuing to hold domestic sovereign claims; possible endogenous increase in home bias via moral suasion in crisis.

### Findings — Sovereign borrowing costs (EMs)
- Home bias is negatively associated with bond yields but with a milder impact relative to AEs (flatter lines in Figure AIII2 (A) and 5th column of Table AIII2).
- Market sentiments play a bigger role in EMs.
  - For a debt level of 60 percent of GDP, the impact of home bias on borrowing cost becomes positive when VIX is slightly above its mean (Figure AIII2(B)).

### Findings — Primary balance adjustments
- Model specification follows Ghosh and others (2011) with squared and cubic terms of lagged debt; estimation by two-step GMM.
- Empirical result: Fiscal policy is less responsive to lagged public debt in AEs and EMs with higher home bias.
  - Panel regression results (2nd column of Table AIII3): primary balance is negatively affected by home bias (0
γ = -0.096** in one specification).
- Interpretation: Given domestic banks’ interest in domestic sovereign claims, sovereigns might be less willing to undertake fiscal consolidation, ceteris paribus.
- Caveat: despite instrumentation, reverse causality may still be an issue.

### Findings — Public debt
- Specifications (3) and (4) analyze concurrent and subsequent debt effects using two-step GMM.
- Empirical result: An increase in home bias in AEs and EMs is associated with a higher public debt level.
  - Table AIII4 highlights significant coefficients: e.g., home bias, lagged, coefficients in combined sample: 2.529*** (IV pooled regression) and 3.753*** (home bias, lagged, in other panels).
- Persistence: subsequent debt level is significantly influenced by home bias due to its persistence.
- Reverse causality caveat: rising public debt might induce moral suasion to place new debt with domestic banks, increasing measured home bias.

### Findings — Debt under distress
- Event study using 17 episodes of debt difficulties (Baldacci and others (2011) and Cruces and Trebesch (2013)).
- Result: AEs and EMs with high home bias tend to experience debt difficulties at a higher level of public debt.
- Example linear regression from Figure AIII3: Debt = 75.05 + 2.63 * Home bias (coefficients reported with significance: (***, 19.25) and (*, 1.37)).

### Robustness
- Main findings broadly hold under robustness checks:
  - Dropping outliers such as Greece and Japan does not change empirical results.
  - Results robust to quantile regression methodologies.
  - Results robust when using alternative home bias measure (domestic sovereign debt as share of total sovereign debt).
- Note: Other studies (Arslanalp and Poghosyan, 2014; Andritzky, 2012) find home bias proxied by holding of domestic sovereign debt as share of total public debt is associated with lower spreads — consistent with a diversification interpretation rather than banks’ home preference.

### Implications and interpretation
- In normal times, higher banks’ home bias can lower sovereign borrowing costs in AEs and EMs, particularly at higher debt levels, likely via reduced expected default incentives.
- In periods of heightened global risk aversion, home bias can cease to lower—and may raise—borrowing costs due to higher risk premia demanded by domestic banks and potential endogenous shifts in portfolio allocations.
- High home bias is associated with weaker fiscal responsiveness to debt and with higher levels of public debt, suggesting potential medium-term fiscal vulnerabilities linked to bank-sovereign home bias.
- Policy-relevant consideration: monitoring bank portfolio composition and the interaction between domestic bank holdings of sovereign debt, fiscal policy responsiveness, and debt dynamics is important, especially under stress.

*Prepared by Tamon Asonuma and Heiko Hesse (both SPR), annex material covering 1999–2012 sample analyses.*

### Annex IV. Market-Based Measures of Bank to

### Annex IV. Market-Based Measures of Bank to Sovereign Contagion

### A. Data Coverage
- Quarterly data over the sample period 2005 Q1–2012 Q4.
- Country groupings:
  - (i) Eurozone countries under stress during the sample period: Greece, Ireland, Italy, Spain and Portugal.
  - (ii) Other eurozone countries: Austria, Belgium, France, Germany and the Netherlands.
  - (iii) Countries outside the eurozone: Sweden, Switzerland, Japan, the U.K. and the U.S.

### B. Construction of the BSC Measure
- BSC defined as the probability of the sovereign defaulting conditional on a median domestic banking default.
- For each country, the top 5 banks by asset size as of end-2012 were included (where data available).
- Steps to construct BSC:
  - Extract marginal default probabilities for the sovereign A and banks B, C, etc., from individual CDS spreads (probabilities estimated by dividing CDS by a recovery rate of 40 percent; altering this value in the range of 25%–50% had little impact on regression results).
  - Estimate joint probabilities P(A,B), P(A,C), P(B,C), ... using the CIMDO methodology (Segoviano, 2006) to obtain the multivariate empirical distribution (CIMDO-distribution) characterizing probabilities of distress and their dependence.
  - Obtain conditional probability P(A/B) via Bayes’ law: P(A/B) = P(A,B)/P(B).
  - For each country, keep the median value of the set {P(A/B), P(A/C), P(A/D)...} as the dependent BSC variable.
- Alternative BSC specifications considered:
  - Equally weighted and probability-weighted approaches (probability weighting = weighting by unconditional probability of bank default).
  - Weighting by relative bank size was infeasible due to lack of consistent balance sheet data.
- Result: dynamics of the contagion indicator remained broadly similar across alternative specifications.

### C. Explanatory Variables (hypothesized determinants of BSC)
- Macroeconomic and fiscal:
  - GDP growth forecast (expected to reduce contagion probability).
  - Short-term public debt/GDP (high stock increases sovereign default risk).
  - Government bonds held by domestic banks (share of banks’ assets) to capture sovereign dependence on domestic bank financing.
- Banking sector characteristics:
  - Bank loans/GDP (scale of banking sector).
  - Banks capital/Assets (capital-to-asset ratio to capture leverage and buffers).
  - Price to book value of banks (valuation prospects).
- Control variables:
  - EZProg: dummy = 1 after the first Greek Program was approved (2010 Q1) to capture intensification of euro debt crisis.
  - OMT: dummy = 1 on and after 2010 Q3 to capture ECB’s “whatever it takes” announcements.
  - MkrPr: market price of risk variable capturing price of risk under distress (constructed from expectations when market price of risk exceeds a threshold).

### D. Baseline Regression Specification
- Panel regression estimated:
  - Pit = β1 X1,it + ... + βn Xn,it + γ1 MkrPrt + γ2 EZProgt + γ3 OMTt + αi + εit
  - Pit denotes BSC; αi are country fixed effects.
- Model choice:
  - Breusch-Pagan test rejected homogeneity (supporting fixed effects).
  - Hausman test rejected random effects.
  - Baseline model uses fixed effects and allows a group-specific slope for price to book value (relationship differs across the three country groups).

### E. Alternative Specifications and Robustness
- Time dummies included in an alternative specification to control for common factors (column 4). All coefficients retain expected sign except price to book for non-eurozone countries.
- Group-specific slopes:
  - Column 5: slope for growth forecast estimated separately by group. Growth forecast not significant for stressed eurozone countries (group 2); significant for non-eurozone countries similar to whole sample.
  - Column 6: government bonds held by domestic banks/Assets most important for stress countries (group 2); not significant for core eurozone or non-eurozone countries.
  - Short-term public debt/GDP less important for countries outside the eurozone.
- Sub-period analysis (pre- and post-first Greek program, columns 8–11):
  - Growth outlook and banking buffers more relevant precrisis.
  - Post-crisis, dependence of government financing on domestic banks and stock of short-term debt became more important.
- Endogeneity robustness:
  - Regressions with explanatory variables lagged by one period show sign, magnitude, and significance unchanged.
  - Intuition: contagion is a low probability event; difficult to justify reverse causality affecting macro and banking fundamentals quantitatively.

### F. Key Estimation Results (selected coefficients from Table AIV1; dependent variable = BSC)
- Model notes: estimation period 2005 Q1 to 2012 Q4; Number of country id = 15. Robust t-statistics in brackets. *** p<0.01, ** p<0.05, * p<0.1.
- Baseline (Panel FE, Column 1) — selected coefficients:
  - Growth forecast: -0.112*** [-3.474]
  - Price to Book Value for Banks * I(group=1): -1.886*** [-11.32]
  - Price to Book Value for Banks * I(group=2): -0.815*** [-6.107]
  - Price to Book Value for Banks * I(group=3): -0.546*** [-4.144]
  - Bank loans/GDP: 1.849*** [6.457]
  - Banks capital/Assets: -6.364* [-1.760]
  - Govt. bonds held by dom. banks/Assets: 8.847*** [3.853]
  - Gvt. Short-Term Debt/GDP: 2.833*** [7.853]
  - Mkt. Price of risk: 17.80*** [8.023]
  - European programs variable (EZProg): 0.334*** [8.429]
  - OMT dummy: -0.501*** [-4.052]
  - R-squared: 0.911
  - Im-Pesaran-Shin p-value: 0.0082
  - Phillips-Perron L* p-value: 0.00405
- Selected alternative/specification highlights (from table columns):
  - Pooled OLS (Column 2): Growth forecast -0.362*** [-12.16]; Mkt. Price of risk 21.07*** [7.412]; OMT dummy -0.904*** [-5.950]; R-squared 0.816.
  - Time dummies (Column 4): Price to Book Value for Banks * I(group=3) = 0.731*** [5.716] (positive sign for non-eurozone in this specification).
  - Group interactions (Column 5): Growth forecast * I(group=1) = -0.230*** [-5.403]; Growth forecast * I(group=3) = -0.190*** [-3.870]; Govt. bonds held by dom. banks/Assets * I(group=2) = 15.24*** [5.481]; Gvt. Short-Term Debt/GDP * I(group=1) = 7.425*** [6.782]; Gvt. Short-Term Debt/GDP * I(group=2) = 4.706*** [10.04]; Gvt. Short-Term Debt/GDP * I(group=3) = 1.156*** [2.931].
  - By period interactions (Columns 8–11): Growth forecast * I(time<2010q1) = -0.178*** [-4.102]; Growth forecast * I(time>2010q1) = -0.0222 [-0.587]; Banks capital/Assets * I(time<2010q1) = -13.10*** [-3.295]; Banks capital/Assets * I(time>2010q1) = -4.593 [-1.316]; Govt. bonds held by dom. banks/Assets * I(time>2010q1) = 8.956*** [3.902]; Gvt. Short-Term Debt/GDP * I(time<2010q1) = 1.938*** [4.983]; Gvt. Short-Term Debt/GDP * I(time>2010q1) = 2.617*** [7.465].
- Overall inference from estimations:
  - Higher market price of risk (Mkt. Price of risk) and higher bank exposures to government (Govt. bonds held by dom. banks/Assets) are associated with higher BSC.
  - Larger banking sector scale (Bank loans/GDP) and greater short-term government debt (Gvt. Short-Term Debt/GDP) are associated with higher BSC.
  - Better bank valuation (higher price to book) is associated with lower BSC across groups (though sign can vary by specification for non-eurozone).
  - OMT announcements are associated with lower BSC (negative and significant OMT dummy).
  - The effects of growth forecast and bank buffers (Banks capital/Assets) were stronger pre-2010 Q1 than post-2010 Q1.

*Prepared by Sheheryar Malik and Miguel Segoviano (MCM); estimation period 2005 Q1–2012 Q4, sample of 15 countries as described above.*

### 4. While there is generally adequate attention to contingent liabilities from explicit

### _122214 - 4. While there is generally adequate attention to contingent liabilities from explicit

### Assessment of contingent liabilities from the financial sector
- 77 percent of respondents has institutions in charge of regular monitoring and quantification of explicit contingent liabilities.
- This information is reported to the ministry of finance in almost all cases.
- 60 percent of respondents (26 countries) indicated that the same monitoring is done for implicit contingent liabilities.
  - A follow-up survey found that only 12 percent (3 countries) actually do this on a regular basis.
  - 35 percent (9 countries) indicated this is done only under special circumstances (for example when stress occurs).
  - The remaining countries did not respond to the follow-up.
- 74 percent of respondents (32 countries) reported the presence of a crisis resolution framework establishing clear procedures and roles for each relevant entity in the face of a banking crisis.
  - More than 60 percent of the countries where this framework is in place (20 countries) indicated that government intervention of the type that could create a liability on the sovereign is explicitly foreseen.
  - In about 40 percent of this group (8 countries), practices for monitoring and quantifying implicit contingent liabilities are not in place.
    - Of these 8 countries:
      - 5 have large banking sectors,
      - 6 have leveraged banking sectors,
      - 3 have high public debt,
      - 2 have a history of banking crises.

### Tools for identifying, measuring, and monitoring risks from the financial sector to the government
- Common risk indicators used in assessments include credit-to-GDP, banks assets to GDP, leverage ratios, and other financial stability indices.
- About two thirds of respondents use the above mentioned indicators to inform discussions with the government on potential risks to their balance sheets from the financial sector (Q20).
  - However, the explicit link to fiscal and debt sustainability is lacking.
  - More than one third of the countries where such discussions do not take place had large and leveraged banking sectors (5 out of 14 countries).
  - All of these countries also indicated that the ministry of finance does not conduct its own analysis of financial sector risks.

### Assessing risks to fiscal and public debt sustainability
- About 60 percent of respondents conduct a public debt sustainability analysis (DSA).
  - In 22 out of 26 countries doing DSAs, the DSA results are used to inform government’s policies.
  - Only 40 percent of respondents indicated the DSA is published.
  - The practice of doing DSAs is more prevalent in countries with high public debt.
- About 60 percent of respondents indicated that stress testing of public debt is a common practice, with the vast majority focused on macroeconomic shocks.
  - Stress testing of public debt is relatively more common among countries with high public debt and/or a history of banking crisis (about 75 percent of each group).
  - Financial sector stress testing is significantly less common, and typically done as shocks to the government cost of borrowing.
  - Even less common is the assessment of the impact of financial sector contingent liabilities on debt even in countries where government intervention is explicitly foreseen in crisis resolution frameworks.
- Cash buffers and coverage of financial sector risks:
  - The practice of holding cash buffers is common across debt managers to help smooth expenditure cycles.
  - About 16 percent of total respondents (7 countries) indicated that estimates of appropriate cash buffers also take into consideration estimates of potential financing needs to cover (partially or fully) a materialization of financial sector risks.

### Key survey response statistics and question-level results
- Survey sent on August 7, 2013 to 117 market access countries; response rate as of October 2013 was 37 percent.
  - Table AV1 regional response rates:
    - AFR: Sent 9, Received 1, Percentage of responses 11.1%
    - APD: Sent 14, Received 5, Percentage of responses 35.7%
    - EUR: Sent 44, Received 26, Percentage of responses 59.0%
    - MCD: Sent 22, Received 4, Percentage of responses 18.2%
    - WHD: Sent 28, Received 4, Percentage of responses 14.3%
    - Total: Sent 117, Received 43, Percentage of responses 36.8%
  - The majority of responses came from advanced economies.
- Table AV2 selected question response counts (Yes / No / N/A / Total):
  - Q3. Is information on financial sector risks shared with the agency responsible for monitoring fiscal and public debt sustainability? — 39 / 3 / 1 / 43 / 91  (Note: table shows "3931439172100" concatenated in source; interpreted as row entries)
  - Q5. Is information sharing specified formally in a memorandum of understanding (MoU) between participating agencies? — 27 / 15 / 14 / 36 / 33 / 52 / 100 (source table entries indicate counts per column)
  - Q7. Are there inter-agency financial stability committees (formal or informal) dedicated to discussing financial sector risks? — 35 / 7 / 14 / 38 / 11 / 62 / 100
  - Q11. Monitoring of explicit contingent liabilities. Are there agency (agencies) responsible for recording, quantifying and reporting financial sector contingent liabilities associated with explicit guarantees to the sector? — 33 / 10 / 0 / 4 / 37 / 72 / 30 / 100
  - Q12. Is the above analysis communicated to the ministry of finance? — 30 / 4 / 9 / 43 / 70 / 92 / 11 / 100
  - Q13. Monitoring of implicit contingent liabilities. Are there agency (agencies) responsible for quantifying and reporting potential financial sector contingent liabilities associated with implicit guarantees? — 26 / 17 / 0 / 4 / 36 / 0 / 40 / 100
  - Q13_a Follow up: how often is this assessment conducted? (regular basis or under special circumstances). Share responding "regular basis" — 3 / 9 / 14 / 26 / 12 / 35 / 54 / 100
  - Q14. Is the above analysis communicated to the ministry of finance? — 22 / 7 / 14 / 43 / 51 / 16 / 33 / 100
  - Q15. In the event of a financial crisis, is there a crisis resolution mechanism in place delineating each agency’s responsibilities in this process? — 32 / 8 / 3 / 43 / 74 / 34 / 37 / 19 / 71 / 100
  - Q16. Is government intervention in banks (of the type that can create a liability on the sovereign) foreseen in this framework? — 20 / 14 / 9 / 43 / 47 / 33 / 21 / 100
  - Q18. Are the above indicators used to inform reports on the health of the financial system? — 40 / 12 / 4 / 39 / 32 / 51 / 00 / 100
  - Q20. Are the above indicators used to inform discussions with the government on potential risks from the financial sector for its budget or balance sheet? — 29 / 10 / 44 / 36 / 72 / 39 / 100
  - Q22. Does the ministry of finance do its own analysis of financial sector risk? — 19 / 20 / 44 / 34 / 44 / 47 / 9 / 100
  - Q24. Does the ministry of finance use analyses conducted by other agencies on financial sector risks? — 33 / 4 / 6 / 43 / 77 / 91 / 41 / 00 / 100
  - Q25. Are stress tests on the evolution of public debt done? — 26 / 16 / 14 / 36 / 37 / 2 / 100
  - Q27. Are there any specific stress tests to assess the impact of financial sector risks on public debt? — 7 / 35 / 14 / 31 / 68 / 12 / 100
  - Q28.a) Is a DSA regularly prepared (e.g., annually)? — 26 / 16 / 14 / 36 / 37 / 2 / 100
  - Q29. Is it published? — 17 / 9 / 17 / 4 / 34 / 02 / 14 / 01 / 00 (source table entries concatenated)
  - Q30–Q31. Which agency is responsible for the production of the DSA? Do other agencies contribute to the production of the DSA? — 15 / 11 / 17 / 43 / 35 / 26 / 40 / 100
  - Q34. Do the structure and conditions of the financial sector affect the design of the public debt management strategy? — 34 / 8 / 14 / 37 / 91 / 92 / 10 / 100
  - Q36. Does the debt manager hold a cash buffer? — 31 / 11 / 14 / 37 / 22 / 62 / 100
  - Q37. Does the debt manager hold a cash buffer explicitly to cover (partially or fully) the materialization of financial sector risks (even if this practice is not publicly disclosed)? — 7 / 34 / 24 / 31 / 67 / 95 / 100
- Note: Table AV2 source indicates "As of September 20, 2013 there were 43 responses excluding China which replied to the survey with written answers to only two questions."

### Policy-relevant observations implied by survey results
- Explicit contingent liabilities are widely monitored and reported to finance ministries; implicit contingent liabilities are much less systematically monitored on a regular basis.
- Crisis resolution frameworks frequently foresee government intervention that can create sovereign contingent liabilities, but monitoring/quantification of implicit liabilities is often absent even where intervention is foreseen.
- Linkages between financial sector risk monitoring and fiscal/public debt sustainability analysis are incomplete: DSAs and stress tests commonly exist but seldom incorporate financial sector contingent liabilities.
- Cash buffers explicitly calibrated to potential financial sector contingent liabilities are uncommon (16 percent; 7 countries).

*International Monetary Fund — From Banking to Sovereign Stress (survey and analysis excerpt).*

### References

### References

### IMF and Working Papers; IMF Staff Notes and Reports
- Abiad A., E. Detragiache, and T. Tressel, 2008, “A New Database of Financial Reforms,” IMF Working Paper WP/08/266, (Washington: International Monetary Fund).
- Abiad, A., G. Dell’Arricia, and B. Li, 2011, “Creditless Recoveries,” IMF Working Paper WP/11/58 (Washington: International Monetary Fund).
- Andritzky, J. R., 2012, “Government Bonds and their Investors: What Are the Facts and Do They Matter?” IMF Working Paper WP/12/158, (Washington: International Monetary Fund).
- Arslanalp, S., and T., Poghosyan, 2014, “Foreign Investor Flows and Sovereign Bond Yields in Advanced Economies,” IMF Working Paper WP/14/27, (Washington: International Monetary Fund).
- Arslanalp, S., and T. Tsuda, 2012, “Tracking Global Demand for Advanced Economy Sovereign Debt,” IMF Working Paper WP/12/284, (Washington: International Monetary Fund).
- Asonuma, T, S. Bakhache, and H. Hesse, 2015, “Is Banks’ Home Bias Good or Bad for Public Debt Sustainability?” IMF Working Paper (Washington: International Monetary Fund).
- Baldacci E., S. Gupta, and C. Mulas-Granados, 2009, “How Effective is Fiscal Policy Response in Systemic Banking Crises?” IMF Working Paper WP/09/160, (Washington: International Monetary Fund).
- Baldacci, E., I. Petrova, N. Belhocine, G. Dobrescu, and S. Mazraani, 2011, “Assessing Fiscal Stress,” IMF Working Paper WP/11/100 (Washington: International Monetary Fund).
- Bologna, P., 2011,“Is There a Role for Funding in Explaining Recent U.S. Banks’ Failures?” IMF Working Paper WP/11/180, (Washington: International Monetary Fund).
- Caceres, C., V. Guzzo, and M. Segoviano, 2010, “Sovereign Spreads: Global Risk Aversion, Contagion, or Fundamentals,” IMF Working Paper WP/10/120, (Washington: International Monetary Fund).
- Claessens, S., and M. A. Kose, 2013, “Financial Crises and Macroeconomic Outcomes: Review and Evidence,” IMF Working Paper WP/13/28, (Washington: International Monetary Fund).
- Claessens, S., A. Kose, L. Laeven, and F. Valencia, 2014, “Financial Crises: Causes, Consequences, and Policy responses,” (Washington: International Monetary Fund).
- Claessens, S., M. A. Kose, M. A., and M. Terrones, 2008, “What Happens During Recessions, Crunches and Busts?” IMF Working Paper WP/08/274, (Washington: International Monetary Fund).
- Claessens, S., C. Pazarbasioglu, L. Laeven, M. Dobler, F. Valencia, O. Nedelescu, and K. Seal, 2011, “Crisis Management and Resolution: Early Lessons from the Financial Crisis,” IMF Staff Discussion Note No. 11/05, (Washington: International Monetary Fund).
- Crowe, C., G. Dell'Ariccia, D. Igan, and P. Rabanal, 2011, “Policies for Macrofinancial Stability: Options to Deal with Real Estate Booms,” IMF Staff Discussion Note No. 11/02 (Washington: International Monetary Fund).
- Dell’Ariccia, G., D. Igan, L. Laeven, and H. Tong, 2012, “Policies for Macrofinancial Stability: How to Deal with Credit Booms,” IMF Staff Discussion Note No. 12/06, (Washington: International Monetary Fund).
- Espinoza, R., and M., Segoviano, 2011, “Probabilities of Default and the Market Price of Risk in a Distressed Economy,” IMF Working Paper WP/11/75, (Washington: International Monetary Fund).
- Gennaioli, N., Martin, A., and S. Rossi, 2014a, “Banks, Government Bonds, and Default: What do the Data Say?” IMF Working Paper WP/14/120. (Washington: International Monetary Fund).
- Gray, D. F., and A. A. Jobst, 2013, “Systemic Contingent Claims Analysis: Estimating Market Implied Systemic Solvency Risk” IMF Working Paper WP/13/54, (Washington: International Monetary Fund).
- Gray, D., M. Gross, J. Paredes, and M. Sydow, 2013, “Modeling Banking, Sovereign, Macro, and Financial Risk in a CCA Global VAR,” IMF Working Paper WP/13/218, (Washington: International Monetary Fund).
- Laeven, L., and F. Valencia, 2008, “Systemic Banking Crises: A New Database,” IMF Working Paper WP/08/224, (Washington: International Monetary Fund).
- Laeven, L., and F. Valencia, 2010, “Resolution of Banking Crises: The Good, the Bad, and the Ugly,” IMF Working Paper WP/10/44, (Washington: International Monetary Fund).
- Laeven, L., and F. Valencia, 2013a, "Systemic Banking Crises Database." IMF Economic Review, Vol. 61, pp. 225–270.
- Laeven, L., L. Ratnovski, and H. Tong, 2014, “Bank Size and Systemic Risk,” IMF Staff Discussion Notes, 14/4, (Washington: International Monetary Fund).
- Lim, C., F. Columba, A. Costa, P. Kongsamut, A. Otani, M. Saiyid, T. Wezel, and X. Wu, 2011, “Macroprudential Policy: What Instruments and How to Use Them? Lessons from Country Experiences,” IMF Working Paper WP/11/238, (Washington: International Monetary Fund).
- Ong, L. L., 2014, “A Guide to IMF Stress Testing: Methods and Models,” (Washington: International Monetary Fund).
- Segoviano, M., and Espinoza, R., 2014 “Determinants of Bank-Sovereign Distress,” IMF Working Paper (forthcoming), (Washington).
- __________, 2006, “Public-Private Partnerships, Government Guarantees, and Fiscal Risk,” (Washington)
- __________, 2007, IMF Manual on Fiscal Transparency, (Washington).
- __________, 2008a, Fiscal Risks: Sources, Disclosure, and Management, (Washington).
- __________, 2008b, World Economic Outlook, October 2008: Financial Stress, Downturns, and Recoveries, World Economic and Financial Surveys (Washington).
- __________, 2009, “Debt Bias and Other Distortions: Crisis-Related Issues in Tax Policy, (Washington).
- __________, 2010, “A Fair and Substantial Contribution by the Financial Sector,” Final report for the G-20, June 2010, (Washington).
- __________, 2011, Global Financial Stability Report, September 2011: “Towards Operationalizing Macroprudential Policies: When to Act?,” (Washington).
- __________, 2011,”The Key Attributes of Effective Resolution Regimes for Financial Institutions— Progress to Date and Next Steps,” (Washington).
- __________, 2013a, “Key Aspects of Macroprudential Policy—Background Paper,” SM/13/145, pp. 33–39, (Washington).
- __________, 2013b, Global Financial Stability Report, October 2013: “Transition Challenges to Stability,” (Washington).
- __________, 2013c, “Nordic Regional Report: Staff Report for the 2013 Cluster Consultation,” IMF Country Report 13/274, (Washington).
- __________, 2013d, Staff Guidance Note for Public Debt Sustainability Analysis in Market Access Countries, (Washington).
- __________, 2014a, Global Financial Stability Report, April 2014: “Moving from Liquidity- to Growth- Driven Markets,” (Washington).
- __________, 2014b, “Cross-Border Bank Resolution—Recent Developments,” IMF Staff Paper (Washington: International Monetary Fund).
- __________, 2014c, Fiscal Monitor, September 2014: “Back to Work: How Fiscal Policy Can Help,” (Washington).
- __________2014d, Global Financial Stability Report, October 2014: “Chapter 2 on Shadow Banking Around the Globe: How Large, and How Risky?” (Washington)

### Journal Articles, Books, and Academic Working Papers
- Acharya V. V., and S. Steffen, 2013, “The Greatest Carry Trade Ever? Understanding Eurozone Bank Risks,” NBER Working Paper No. 19039, (Cambridge: National Bureau of Economic Research).
- Allen, F., and D. Gale, 2004, “Financial Fragility, Liquidity, and Asset Prices,” Journal of European Economic Association, Vol. 2(6), pp. 1015–48.
- Angeloni, C., and G. Wolff, 2012, “Are Banks Affected by their Holdings of Government Debt?” Bruegel Working Paper 2012/07, (Brussels: Bruegel).
- Ardagna, S., Caselli, F., and T. Lane, 2007, “Fiscal Discipline and the Cost of Public Debt Service: Some Estimates for OECD Countries,” The B. E. Journal of Macroeconomics, Vol. 7(1), pp. 1–35.
- Balteanu, I., A. Erce, and L. Fernandez, 2012, “Bank Crises and Sovereign Defaults: Exploring the Links,” manuscript, Bank of Spain.
- Bank of International Settlement, 2006, “International Convergence of Capital Measurement and Capital Standards,” (Basel)
- Bolton P., and O. Jeanne, 2011, "Sovereign Default Risk and Bank Fragility in Financially Integrated Economies," IMF Economic Review, Vol. 59(2).
- Brixi, H. P., and A. Schick, 2002, “Government at Risk: Contingent Liabilities and Fiscal Risks,” (Washington and New York: World Bank and Oxford University Press).
- Broner, F., A. Erce, A. Martin, and J. Ventura, 2014, “Sovereign Debt Markets in Turbulent Times: Creditor Discrimination and Crowding-out Effects,” Journal of Monetary Economics, Vol. 61, pp. 114–42.
- Calomiris, C., D. Klingebiel, and L. Laeven, 2005, “A Taxonomy of Financial Crisis Restructuring Mechanisms: Cross-Country Experience and Policy Implications,” eds. by P. Honohan and L. Laeven, “Systemic Financial Crises: Containment and Resolution,” (Cambridge University).
- Cecchetti, S.G., M. Kohler, and C. Upper, 2009, “Financial Crises and Economic Activity,” NBER Working Paper No. 15379 (Cambridge: National Bureau of Economic Research).
- Chitu, L., 2013, “Was Unofficial Dollarisation/Euroisation an Amplifier of the ‘Great Recession’ of 2007–09 in Emerging Economies?” Comparative Economic Studies, Vol. 55, pp. 233–65.
- Chinn, M. D., and H. Ito, 2006, "What Matters for Financial Development? Capital Controls, Institutions, and Interactions," Journal of Development Economics, Vol. 81 (1), pp. 163–92.
- Claessens, S., and D. Klingebiel, 2002, “Fiscal Risks of the Banking System: Approaches to Measuring and Managing Contingent Government Liabilities in the Banking Sector,” eds. by Hana Polackova Brixi and Allen Schick, “Government at Risk. Contingent Liabilities and Fiscal Risk,” April 2002, pp. 311–34 (World Bank and Oxford University Press).
- Claessens, S., D. Klingebiel and L. Laeven, 2003, “Financial Restructuring in Banking and Corporate Sector Crises, What Policies to Pursue?,” eds by M. Dooley and J. Frankel, “Currency Crises Management,” pp. 147–80 (NBER and University of Chicago Press).
- Claessens, S., D. Klingebiel and L. Laeven, 2005, “Crisis Resolution, Policies, and Institutions: Empirical Evidence,” eds by P. Honohan and L. Laeven, “Systemic Financial Distress: Containment and Resolution,” Chapter 6, pp. 169 –96 (Cambridge: Cambridge University Press).
- Claessens, S., and M. A. Kose, 2013, “Financial Crises and Macroeconomic Outcomes: Review and Evidence,” IMF Working Paper WP/13/28, (Washington: International Monetary Fund).
- Claessens, S., M. A. Kose, M. A., and M. Terrones, 2008, “What Happens During Recessions, Crunches and Busts?” IMF Working Paper WP/08/274, (Washington: International Monetary Fund).
- Claessens, S., 2011, “Financial Cycles: What? How? When?,” NBER International Seminar on Macroeconomics 2010, eds by R. H. Clarida, and F. Giavazzi, Vol.7, pp. 303–44.
- Claessens, S., 2012, “How Do Business and Financial Cycles Interact?” Journal of International Economics, Vol. 87, No.1, pp. 178–90.
- Claessens, S., C. Pazarbasioglu, L. Laeven, M. Dobler, F. Valencia, O. Nedelescu, and K. Seal, 2011, “Crisis Management and Resolution: Early Lessons from the Financial Crisis,” IMF Staff Discussion Note No. 11/05, (Washington: International Monetary Fund).
- Cruces, J., and C. Trebesch, 2013, “Sovereign Defaults: The Price of Haircuts,” American Economic Journal: Macroeconomics, Vol.5(3), pp. 85–117.
- Davies, M., 2011, “The Rise of Sovereign Credit Risk: Implications for Financial Stability,” BIS Quarterly Review.
- Dell’Ariccia, G., G. E. Detragiache, and R. Rajan, 2008, “The Real Effect of Banking Crises,” Journal of Financial Intermediation, Vol. 17, pp. 89–112.
- De Marco, F., and M. Macchiavelli, 2014, “The Political Origin of Home Bias: the Case of Europe,” manuscript, Boston College.
- De Mooij, R., 2011, “Tax Biases to Debt Finance: Assessing the Problem, Finding Solutions,” IMF Staff Discussion Note No.11/11, (Washington: International Monetary Fund).
- De Mooij, R., Keen, M. and M., Orihara, 2013, “Taxation, Bank Leverage, and Financial Crises”, eds. by R. de Mooij and G. Nicodeme, “Taxation and Regulation of the Financial Sector,” pp. 229–52, MIT Press.
- Demirgüç-Kunt, A., and E. Detragiache, 1998, “The Determinants of Banking Crises in Developing and Developed Countries,” IMF Staff Papers, Vol. 45, pp. 81–109.
- Demirgüç-Kunt A., and E., Detragiache, 2010, “Basel Core Principles and Bank Risk,” IMF Working Paper WP/10/81, (Washington: International Monetary Fund).
- Detragiache, E., and G. Ho, 2010, “Responding to Banking Crises: Lessons from Cross-Country Evidence,” IMF Working Paper WP/10/18 (Washington: International Monetary Fund).
- Deutsche Bank, 2013, “Financial Crises: Past and Present,” Global Economic Perspectives, Deutsche Bank Market Research, (15 May).
- Devereux, M., N. Johannesen, and J. Vella, 2013, “Can Taxes Tame the Banks? Evidence from European Bank Levies,” Working Papers 1325, Oxford University Centre for Business Taxation.
- European Commission, 2012, "Fiscal Sustainability Report 2012," European Economy 8/2012.
- Ghosh, A. R., Kim, J. I., Mendoza, E. G., Ostry, J. D., and M. Qureshi, 2011, “Fiscal Fatigue, Fiscal Space and Debt Sustainability in Advanced Economies,” NBER Working Paper No. 16782.
- Gray, D. F., and A. A. Jobst, 2013, “Systemic Contingent Claims Analysis: Estimating Market Implied Systemic Solvency Risk” IMF Working Paper WP/13/54, (Washington: International Monetary Fund).
- Gray, D., M. Gross, J. Paredes, and M. Sydow, 2013, “Modeling Banking, Sovereign, Macro, and Financial Risk in a CCA Global VAR,” IMF Working Paper WP/13/218, (Washington: International Monetary Fund).
- Harding, D., and A. Pagan, 2002, “Dissecting the Cycle: A Methodological Investigation,” Journal of Monetary Economics, Vol. 49, pp. 365–81.
- Hoggarth, G., R. Reis, and V. Saporta, 2002, “Output Costs of Banking System Instability: Some Empirical Evidence,” Journal of Banking and Finance, Vol. 26, pp. 825–55.
- Honohan, P., and D. Klingebiel, 2003, “The Fiscal Cost Implications of an Accommodating Approach to Banking Crises,” Journal of Banking and Finance, Vol.27, pp. 1539–60.
- Hoshi, T, A. Kashyap, and R. Cabellero, 2013, "Zombie Lending and Depressed Restructuring in Japan," manuscript, University of Chicago Graduate School of Business.
- Kroszner, R., L. Laeven, and D. Klingebiel, 2007, “Banking Crises, Financial Dependence, and Growth,” Journal of Financial Economics, Vol. 84, pp. 187–228.
- Laeven, L., and Fabián Valencia, 2013b, “The Real Effects of Financial Sector Interventions during Crises,” Journal of Money, Credit, and Banking, Vol. 45(1), pp. 147–77.
- Levy-Yeyati, E., 2006, “Financial Dollarization: Evaluating the Consequences,” Economic Policy, Vol. 21(4–5), pp. 61–118.
- Luo, L., 2005, “Bootstrapping Default Probability Curves,” Journal of Credit Risk, Vol.1 (2), (Fall 2005).
- Mauro, P., Romeu, R., Binder, A., and A. Zaman, 2013 “A Modern History of Fiscal Prudence and Profligacy,” IMF Working Paper WP/13/5, (Washington: International Monetary Fund).
- Merler, S., and J., Pisani-Ferry, 2012, “Hazardous Tango: Sovereign-Bank Interdependence and Financial Stability in the Euro Area,” Banque de France Financial Stability Review No. 16, April 2012.
- Reinhart, C. M., and K. S. Rogoff, 2011, “From Financial Crash to Debt Crisis,” American Economic Review, Vol. 101, pp. 1676–06.
- Reinhart, C., and K. Rogoff, 2014, “Recovery from Financial Crises: Evidence from 100 Episodes,” forthcoming in: American Economic Review Papers and Proceedings.
- Reinhart C., and M. S. Sbrancia, 2011, “The Liquidation of Government Debt,” NBER Working Paper No. 16893, (Cambridge: National Bureau of Economic Research).
- Schuknecht, L., 2010, “Fiscal Activism in Booms, Busts and Beyond,” Occasional Paper, Bank of Italy.
- Segoviano, M., 2006, “The Conditional Probability of Default Methodology,” Financial Markets Group, London School of Economics, Discussion Paper No. 558.
- Taylor, A., 2012, “The Great Leveraging,” NBER Working Paper No. 18290, (Washington).

### Country Documents, Central Bank / Regulatory, Industry Reports, Manuscripts, and Other Sources
- Acharya V. V., and S. Steffen, 2013, “The Greatest Carry Trade Ever? Understanding Eurozone Bank Risks,” NBER Working Paper No. 19039, (Cambridge: National Bureau of Economic Research).
- Balteanu, I., A. Erce, and L. Fernandez, 2012, “Bank Crises and Sovereign Defaults: Exploring the Links,” manuscript, Bank of Spain.
- Davies, M., 2011, “The Rise of Sovereign Credit Risk: Implications for Financial Stability,” BIS Quarterly Review.
- Devereux, M., N. Johannesen, and J. Vella, 2013, “Can Taxes Tame the Banks? Evidence from European Bank Levies,” Working Papers 1325, Oxford University Centre for Business Taxation.
- Deutsche Bank, 2013, “Financial Crises: Past and Present,” Global Economic Perspectives, Deutsche Bank Market Research, (15 May).
- Ireland, 2014, Draft National Risk Assessment, Department of the Taoiseach.
- Italy, 2012, Economic and Financial Document, Ministry of Economy and Finance.
- The Philippines, 2012, “Fiscal Risks Statement,” Development Budget Coordination Committee.
- U.S. Financial Stability Oversight Council (FSOC), 2014, “Annual Report.”

*From Banking to Sovereign Stress — INTERNATIONAL MONETARY FUND*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_122214.pdf_