## Managing the Sovereign-Bank Nexus — Executive Summary

## Source details

**Canonical URL:** [Managing the Sovereign-Bank Nexus — Executive Summary](https://www.imf.org/-/media/files/publications/dp/2018/45133-dp1816-managing-the-sovereign-bank-nexus.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/dp/2018/45133-dp1816-managing-the-sovereign-bank-nexus.pdf.md)
- [Structured JSON version](/-/media/files/publications/dp/2018/45133-dp1816-managing-the-sovereign-bank-nexus.pdf.json)

---

### Overview and purpose
- The financial health of banks and sovereigns is intertwined in a “sovereign-bank nexus” that may multiply and accelerate vulnerabilities in each sector, and lead to adverse feedback loops.
- The paper identifies the channels forming the nexus, assesses their empirical relevance, and discusses policy implications to reduce the likelihood of severe stress and lower the potency of the nexus.

### Main conclusions
- Banks and sovereigns are linked by multiple interacting channels:
  - (1) the sovereign-exposure channel (banks hold large amounts of sovereign debt),
  - (2) the safety net channel (banks are protected by government guarantees),
  - (3) the macroeconomic channel (the health of banks and governments affect and is affected by economic activity).
  - Evidence suggests that all three channels are relevant.
- Policies aimed at weakening the nexus should be designed from a holistic point of view; measures targeting one channel may have undesired consequences for others.
- The nexus can be weakened but not completely severed; policies should be designed acknowledging this constraint.
- Strengthening balance sheets and governance of banks and sovereigns reduces the nexus’s relevance:
  - Larger fiscal buffers and better management of public debt improve debt sustainability and reduce sovereign-related bank distress.
  - Larger capital buffers and better prudential frameworks strengthen banks and reduce bank-induced sovereign distress.

### Key policy directions (high level)
- Monitor and limit excessive domestic sovereign concentration in bank portfolios.
- Strengthen resolution frameworks and implement measures to reduce implicit guarantees (for example, higher capital for systemic institutions, recovery and resolution planning, TLAC/bail-inable debt).
- Calibrate measures to preserve legitimate market functions of sovereign bond holdings while reducing excessive concentration.
- Enhance transparency and disclosure of sovereign holdings by jurisdiction, currency, and regulatory treatment.
- Improve cross-border resolution arrangements and cooperation.

### Evidence on co-occurrence of crises
- Banking crisis conditional on sovereign debt crisis: 22.30%
- Sovereign debt crisis conditional on banking crisis: 51.00%
- Note: Shares computed for 2000–14 across 66 countries.

---

### Channels of the sovereign-bank nexus

- Sovereign-exposure channel
  - Banks demand and hold sovereign debt for liquidity management, credit exposure, market-making, collateral, and central bank operations.
  - Banks are directly exposed to sovereign risk and are an important source of financing for the government.
- Safety net channel
  - Central bank and government backstops create implicit and explicit government guarantees that strengthen the nexus.
  - An increase in sovereign risk lowers the government’s ability to provide a backstop; banking crises activate costly guarantees and resolution policies with negative fiscal effects.
- Macroeconomic channel
  - Financial instability and banking crises impair credit provision and monetary transmission, worsen economic activity, and deteriorate fiscal accounts.
  - Sovereign risk can prompt fiscal consolidation and higher funding costs, contracting economic activity and worsening banks’ loan portfolios.

---

### Sovereign-exposure channel — quantitative evidence

- Cross-country averages (Bankscope, 1999–2014; 858 banks from 46 countries):
  - Advanced economies: average exposure ranged from 6.0 percent to 9.1 percent of bank assets.
  - Emerging and developing economies (EMDEs): ranged from 15.6 percent to 20.9 percent of total assets.
- Selected Table 2 figures:
  - 1999 mean exposure: All 8.1, AEs 6.4, EMDEs 17.7; median: All 4.8, AEs 4.2, EMDEs 15.7
  - 2005–14 mean exposure: All 9.1, AEs 7.8, EMDEs 16.8; median: All 6.3, AEs 5.5, EMDEs 16.7
- Illustrative capital impact:
  - For a hypothetical bank with a leverage ratio of 6.6 percent (assets 15 times capital) and a sovereign exposure of 10 percent of assets, a 10 percent loss on sovereign bonds would imply a 15 percent reduction in bank capital.
- Sovereign restructuring/haircut simulations (Table 6, 2005–13 data):
  - [1] Sovereign exposure/Total assets means — All 9.57, AEs 7.7, EMDEs 15.9; medians — All 6.5, AEs 5.4, EMDEs 14.0.
  - [2] Comm equity/Total assets means — All 7.1, AEs 6.4, EMDEs 10.6; medians — All 6.1, AEs 5.8, EMDEs 9.3.
  - Loss on sovereign exposure after haircut expressed as percent of total assets:
    - 37% haircut means means — All 3.5, AEs 2.9, EMDEs 5.9; medians — All 2.4, AEs 2.0, EMDEs 5.2.
    - 50% haircut means means — All 4.7, AEs 3.9, EMDEs 8.0; medians — All 3.2, AEs 2.7, EMDEs 7.0.
  - Loss on sovereign exposure after haircut as percent of common equity:
    - 37%: Means — All 49.5, AEs 44.8, EMDEs 55.8; medians — All 39.2, AEs 34.2, EMDEs 55.4.
    - 50%: Means — All 66.9, AEs 60.5, EMDEs 75.5; medians — All 53.0, AEs 46.3, EMDEs 74.9.

### Determinants of bank sovereign holdings (country-year regressions, 1981–2014)
- Selected coefficient estimates (dependent variable: banks’ government debt holdings % of banking sector assets):
  - T-bill rate: 0.0982** (0.040) in (1); 0.1588*** (0.058) in (2); 0.1194** (0.055) in (3).
  - Real GDP growth: 20.0056 (0.085) in (1); 0.1128 (0.076) in (2); 0.0518 (0.075) in (3).
  - Inflation (eop): 0.0166 (0.030) in (1); 20.0604 (0.053) in (2); 20.0420 (0.056) in (3).
  - NER (% change, eop): 20.0309* (0.016) in (1); 20.0073 (0.020) in (2); 20.0096 (0.022) in (3).
  - Public debt (% GDP): 0.0644** (0.030) in (1); 0.0773** (0.035) in (2); 0.0751** (0.031) in (3).
  - Credit/GDP: 20.0863*** (0.031) in (2).
  - Stock market capitalization (% GDP): 20.0319*** (0.011) in (2).
  - Number of listed companies (per 1,000,000 people): 20.0291* (0.015) in (3).
- Model fit and coverage:
  - Observations: 1,763 in (1); 954 in (2); 1,002 in (3).
  - R-squared: 0.723 in (1); 0.789 in (2); 0.775 in (3).
  - Number of countries: 94 in (1); 70 in (2); 70 in (3).
- Interpretation: results support the liquidity view — banks hold more government debt during periods of high interest rates and in countries with lower private sector credit-to-GDP ratios; less developed financial systems correlate with higher government debt holdings.

### Behavior during sovereign distress and euro area evidence
- Around sovereign crises (53 advanced and emerging economies, 1970–2014):
  - On average, banks do not change their exposure during periods of sovereign distress.
- Euro area (late 2007 to mid-2013):
  - ECB series: Sep-07 total 4.2 → Jun-13 total 6.5 (percent of assets).
  - Bankscope mean total 2007–2013: 5.1 (percent of assets).
  - Domestic debt holdings drove the increase; foreign sovereign holdings remained ~1.5 percent of assets.
  - Distressed countries (Greece, Ireland, Italy, Portugal, Spain): domestic exposure increased from 2.5 percent to 7 percent of assets.
  - Non-distressed core countries had smaller increases.
- Interpretation: disproportionate rise in domestic holdings in distressed countries is consistent with risk-taking and financial repression; limited increase in foreign sovereign holdings suggests liquidity alone may not explain the behavior.

---

### Safety net channel — empirical findings and fiscal consequences

- Deposit rates and sovereign CDS (sample 2006–13, half-year frequency, eight euro area countries):
  - Sovereign CDS coefficients: 0.5408*** (0.119) in column (1); 0.5033*** (0.141); 0.4892*** (0.129); 0.4755*** (0.119).
  - Bank CDS: 0.0777*** (0.020) in column (1); 0.0700*** (0.017) in column (2).
  - Interpretation: sovereign CDS is a significant determinant of the deposit rate paid by banks, even after controlling for bank CDS; relationship stronger for weaker banks and weaker sovereigns.
- Event-study (EBA stress test releases):
  - Interaction Sovereign CDS * Domestic sovereign exposure on banks’ stock returns: −1.664*** (0.396) in column (1) (others similar).
  - In CDS-spread regressions, interaction implies: a bank with 10 percent of domestic sovereign bond holdings has a CDS spread higher by 44.3 basis points than a bank with zero such holdings (column 4).
- Fiscal costs of systemic banking crises (1970–2011) — medians (Table 10):
  - Output loss: All 23.2 percent of GDP; Advanced 32.4; Emerging 33.6; Developing 0.7.
  - Increase in public debt: All 12.1 percent of GDP; Advanced 23.6; Emerging 9.1; Developing 10.9.
  - Direct fiscal costs: All 6.8 percent of GDP and 12.5 percent of financial system assets; Advanced 4.2 percent of GDP and 2.1 percent of financial system assets; Emerging 8.3 percent of GDP and 21.3 percent of financial system assets; Developing 10.0 percent of GDP and 18.3 percent of financial system assets.
  - Note: Extremes include Indonesia >50 percent of GDP (1997–2001) and Iceland and Ireland >30 percent of GDP in recent crisis episodes.

---

### Macroeconomic channel — impact on growth and fiscal dynamics
- Banking crises are associated with large real effects:
  - Average cumulative output loss associated with banking crises: about 23 percent of GDP.
  - Banking crises act as multipliers of other negative shocks and produce deeper and longer recessions.
- Banking expansion indicators:
  - Banking expansion defined as five-year cumulative change in private-sector credit-to-GDP above median; "severe" if in top quartile.
  - Median recovery time for recessions preceded by severe banking expansions is more than twice that for recessions preceded by nonsevere expansions.
- Fiscal dynamics:
  - Primary balances tend to be higher during banking expansions but deteriorate more quickly during ensuing downturns; fiscal sustainability pressures last longer in banking recessions.

---

### Policy implications and recommended measures (preserving trade-offs)

- Sovereign-exposure channel — recommended regulatory measures
  - Impose a positive but time-invariant (through-the-cycle) risk weight for domestic sovereign exposures to reduce the incentive to invest excessively in domestic sovereign bonds while avoiding regulatory-induced procyclicality.
  - Calibrate capital requirements that increase with sovereign concentration on banks’ balance sheets.
  - Consider capital surcharges reflecting concentrations of sovereign holdings to discourage home bias and incentivize a more diverse pool of liquid assets.
  - Consider replacing domestic sovereign exposures with securities backed by a diversified pool of assets across countries (sovereign-bond-backed securities).
  - Treat exposures to central banks more leniently than central governments, subject to strong governance for monetary policy.
  - Narrow favorable regulatory treatment for loans to subnational governments and public sector entities that do not serve the same liquidity/collateral role as central government bonds.
  - Increase transparency and disclosure of sovereign exposures by jurisdiction, currency denomination, and account classification.

- Safety net channel — recommended measures and caveats
  - Weaken the safety net channel by restricting and reducing government guarantees and introducing private loss-sharing arrangements in bank resolution.
  - Implement ex-ante bail-in policies that impose losses on shareholders and uninsured debtholders to minimize moral hazard and enhance market discipline.
  - Strengthen ex-post resolution frameworks to limit recourse to public funds, facilitate speedy resolution, preserve bank value, and minimize spillovers.
  - Maintain constrained flexibility to allow temporary public sector support in times of severe contagion risk to mitigate system-wide distress.
  - Prioritize full and timely implementation of reforms: robust early intervention powers, effective resolution toolkits and adequate loss-absorbing capacity, effective cross-border cooperation, and credible recovery and resolution plans.

- Cross-border integration — considerations
  - Cross-border diversification can weaken domestic macro and sovereign exposure channels but may strengthen safety-net channel because banks become larger and resolution remains national in practice.
  - Intermediate holding company requirements can facilitate local resolvability but may reinforce links between subsidiaries and sovereign hosts.
  - A fully integrated safety net (for example, a banking union) could materially weaken domestic nexus channels via centralized deposit insurance, lender-of-last-resort functions, supervision, and resolution.

- Trade-offs and design principles
  - Policies must be designed holistically; measures targeting one channel may have undesired consequences for others (for example, higher risk weights can be procyclical and tighten credit in downturns).
  - Recognize that some sovereign bond holdings are natural and useful for liquidity, collateral, and market functioning; calibration must weigh benefits and costs.
  - Improved disclosure of sovereign holdings and market discipline can complement prudential measures.

---

### Regulatory treatment summary (Basel and practice)
- Under the Basel standardized approach, sovereign exposures are risk-weighted according to external ratings on an increasing scale; national discretion may apply a lower weight to domestic sovereign debt provided it is denominated and funded in domestic currency.
- All Basel Committee member jurisdictions apply a zero risk weight to domestic sovereign debt in domestic currency; this exemption is extended across the European Union.
- Risk Weights for Sovereign Exposures in the Standardized Approach (Percent):
  - Credit rating: AAA to AA– | Risk weight: 0
  - Credit rating: A+ to A– | Risk weight: 20
  - Credit rating: BBB+ to BBB– | Risk weight: 50
  - Credit rating: BB+ to B– | Risk weight: 100
  - Credit rating: Below B– | Risk weight: 150
  - Credit rating: Unrated | Risk weight: 100
- Other regulatory features noted:
  - Sovereign exposures often face zero or low risk weights, no large-exposure/concentration limits, inclusion in leverage ratio denominator, and favorable treatment under Liquidity Coverage Ratio as high-quality liquid assets.
  - No Pillar 1 charge for interest rate risk on sovereign debt; enhancements pursued under Pillar 2.
  - Some jurisdictions perform stress tests including sovereign shocks; results can guide Pillar 2 supervisory actions.

*Source: Managing the Sovereign-Bank Nexus — Executive Summary and Section IV (safety net channel, sovereign exposure channel, macroeconomic channel).*

### Executive Summary ������������������������������������������������������������������������������������������������������

### Managing the Sovereign-Bank Nexus — Executive Summary

### Overview and purpose
- The financial health of banks and sovereigns is intertwined in a “sovereign-bank nexus” that may multiply and accelerate vulnerabilities in each sector, and lead to adverse feedback loops.
- The paper identifies the channels forming the nexus, assesses their empirical relevance, and discusses policy implications to reduce the likelihood of severe stress and lower the potency of the nexus.

### Main conclusions
- Banks and sovereigns are linked by multiple interacting channels:
  - (1) the sovereign-exposure channel (banks hold large amounts of sovereign debt),
  - (2) the safety net channel (banks are protected by government guarantees),
  - (3) the macroeconomic channel (the health of banks and governments affect and is affected by economic activity).
  - Evidence suggests that all three channels are relevant.
- Policies aimed at weakening the nexus should be designed from a holistic point of view; measures targeting one channel may have undesired consequences for others.
- Because of the systemic nature of banks and sovereigns, the nexus can be weakened but not completely severed; policies should be designed acknowledging this constraint.
- Stronger balance sheets and governance of banks and sovereigns will reduce the nexus’s relevance:
  - Larger fiscal buffers and better management of public debt improve debt sustainability and reduce the risk of sovereign-related bank distress.
  - Larger capital buffers and better prudential frameworks strengthen banks and reduce the risk of bank-induced sovereign distress.
- Policies that discourage banks from holding excessive sovereign bonds (for example, positive risk weights or limits on exposures) can improve financial stability and market efficiency, but should be designed to minimize procyclical effects.
  - Banks hold some sovereign bonds as a natural feature of the financial system; calibration should consider benefits and costs of smaller holdings.
  - Additional disclosure of sovereign holdings would strengthen market discipline.
- Limits on public guarantees and private loss-sharing arrangements for bank resolution may reduce excessive risk taking (ex ante) and the direct fiscal cost of bank resolution (ex post).
  - Efforts to “end too-big-to-fail” go in the right direction.
  - Simply limiting government backstops and safety nets could worsen an eventual banking crisis and increase its indirect fiscal and economic costs.
  - Reforms of safety net arrangements should start with a sound resolution framework with broad resolution powers and tools, effective cross-border cooperation, and robust early intervention powers.
- There is an international dimension to the nexus:
  - The nexus would be weakened if banks were fully diversified across countries and had access to a supra-national safety net.
  - Because a supra-national safety net is missing, cross-border diversification should not lead to complacency as bank exposures can change quickly during crises.
  - The lack of effective arrangements for cross-border resolution complicates the situation.

### Evidence on co-occurrence of crises
- Table 1. Banking Crises and Sovereign Distress
  - Sovereign debt crisis, conditional on banking crisis: 51.00%
  - Banking crisis, conditional on sovereign debt crisis: 22.30%
  - Note: The table depicts the share of crisis-years identified as a banking crisis or sovereign debt crisis, conditional on a banking crisis or sovereign debt crisis occurring, respectively, during 2000–14 for 66 countries.

### Channels of the sovereign-bank nexus (identified)
- The sovereign-exposure channel:
  - Banks demand and hold large amounts of sovereign debt for liquidity management, credit exposure, market-making, and other purposes.
  - Banks are directly exposed to sovereign risk and are an important source of financing for the government.
- The safety net channel:
  - The banking system operates against the background of safety net arrangements that support financial stability.
  - Central bank and government backstops create implicit and explicit government guarantees that strengthen the nexus.
- The macroeconomic channel:
  - Banks provide credit to households and firms; financial instability and banking crises can have a large impact on real economic activity, worsening fiscal accounts.
  - The banking system is an important channel of transmission for monetary policy; its impairment can put an undue burden on fiscal policy.

### Regulatory and crisis-management influences
- Regulatory frameworks and crisis management policies significantly influence the nexus:
  - Policies that favor sovereign bond holdings, ineffective resolution schemes, and inadequate treatment of systemically important institutions strengthen the nexus.
  - Post-global financial crisis reforms have aimed at reducing severe financial and fiscal stress and lowering the sovereign-bank amplification mechanism, including:
    - Increasing capital, liquidity, and leverage requirements.
    - Introducing long-term unsecured debt instruments that can be written down or converted into equity in resolution.
    - Strengthening macroprudential frameworks to contain systemic risk.
    - Enhancing crisis management tools and improving resolution frameworks for systemic banks.
  - The Basel Committee reviewed regulatory treatment of sovereign exposures and policy options (such as positive risk weights and exposure limits) but could not reach consensus on introducing these measures (Basel Committee on Banking Supervision 2017).

### Trade-offs and policy design considerations
- Measures aimed at a single channel may have undesired side effects on other channels; for example:
  - Restrictions on a government’s ability to support financial institutions may limit the direct exposure of the sovereign to bank distress but could exacerbate overall banking distress through spillovers and worsen macro effects, indirectly hurting fiscal accounts.
- Effective reforms require:
  - Holistic design that accounts for interactions among channels.
  - Recognition that the nexus can be weakened but not fully eliminated.
  - Calibration to preserve legitimate market functions of sovereign bond holdings while reducing excessive concentration.
  - Improved disclosure and market discipline for sovereign holdings.
  - Robust resolution frameworks, cross-border cooperation, and early intervention to reform safety nets without increasing crisis costs.

*Source: Executive Summary, Managing the Sovereign-Bank Nexus.*

### Section IV discusses this channel.

### 45133-dp1816-managing-the-sovereign-bank-nexus - Section IV discusses this channel.

### The safety net channel
- Safety net arrangements and backstops provided by central banks and sovereigns generate spillovers from bank to sovereign risk and vice versa.
- Mechanisms:
  - An increase in sovereign risk lowers the government’s ability to provide a backstop, thereby hurting banks.
  - Banking crises activate backstops, guarantees, and other costly resolution policies with negative effects on the fiscal accounts.
  - In some jurisdictions, the significant role of state-owned banks strengthens this channel.
- Reference: "Section V provides an overview."

### The macroeconomic channel
- Increases in sovereign risk have contractionary effects on economic activity due to:
  - the associated need for fiscal consolidation,
  - higher funding costs throughout the economy,
  - the impact on policy uncertainty.
- Weaker economic activity negatively affects banking system stability through likely deterioration of banks’ loan portfolios.
- The channel also runs in reverse: banking crises impair monetary policy transmission and thus government finances.
- Reference: "This channel is discussed in Section VI."

### Interaction of channels and feedback loops
- Channels operate simultaneously and interact. Example feedback loop described:
  - Deterioration of a country’s fiscal position and creditworthiness may reduce sovereign bond prices → generates losses and weakens banks’ capital through their sovereign holdings → undermines credit provision → lowers economic activity → further deteriorates fiscal position.
  - Adverse shocks to banks’ balance sheets may change banks’ demand for sovereign bonds:
    - Banks may reduce risk exposures and increase the relative weight of sovereign bonds.
    - Or overall portfolios may shrink with negative consequences for fiscal balances and the real economy.
- Empirical objective: provide direct evidence of whether each channel is empirically relevant rather than precisely apportioning relative contributions.

### The sovereign exposure channel — why banks hold government debt
- Rationale categories (non-mutually-exclusive):
  - Liquidity:
    - Sovereign paper is relatively safe, usually sizable and active, provides diversification, and serves as the benchmark for pricing other securities.
    - Sovereign instruments are widely accepted collateral, support payment systems, hedging, and broader market operations.
    - Central bank liquidity operations are typically conducted extensively through government debt.
    - Current regulatory framework favors domestic sovereign holdings (e.g., zero risk weights on domestic government bonds in local currency under the Basel Committee standardized approach).
    - Liquidity Coverage Ratio criteria frequently narrow eligible securities, leading to a large share of sovereign debt.
  - Risk taking:
    - Banks may not fully price sovereign risk because they expect bailouts in the event of sovereign default.
    - Limited liability and risk shifting can lead banks to "bet" on risky government debt correlated with states where they would be more likely to fail.
  - Financial repression:
    - Governments may implicitly or explicitly encourage or force banks to hold sovereign debt (historical examples post-World War II and evidence of moral suasion).

### How much public debt do banks hold? (evidence and implications)
- Cross-country averages (Bankscope, 1999–2014):
  - Advanced economies: average bank exposure to government debt ranged from 6.0 percent to 9.1 percent of bank assets.
  - Emerging and developing economies (EMDEs): ranged from 15.6 percent to 20.9 percent of total assets.
- Table 2 (selected figures reported in source):
  - 1999 mean exposure: All 8.1, AEs 6.4, EMDEs 17.7; median: All 4.8, AEs 4.2, EMDEs 15.7
  - 2005–14 mean exposure: All 9.1, AEs 7.8, EMDEs 16.8; median: All 6.3, AEs 5.5, EMDEs 16.7
  - Note: Estimates reported for 858 banks from 46 countries (732 banks from advanced economies and 126 banks from EMDEs).
- Illustrative capital impact:
  - For a hypothetical bank with a leverage ratio of 6.6 percent (assets 15 times capital) and a sovereign exposure of 10 percent of assets, a 10 percent loss on sovereign bonds would imply a 15 percent reduction in bank capital.
- Sovereign bonds often carry zero to low risk weights, so banks may assign little or no capital against such potential losses.

### How sovereign exposure varies over time and across countries
- Liquidity view predictions:
  - Banks should hold more government debt when fewer lending opportunities exist and private alternatives are scarce.
  - Sovereign bond holdings can act as a buffer when liability-side adjustments are slow.
- Empirical determinants (country-year regressions, 1981–2014; dependent variable: banks’ government debt holdings in percent of total banking sector assets):
  - Selected coefficient estimates from Table 3 (standard errors in parentheses):
    - T-bill rate: 0.0982** (0.040) in column (1); 0.1588*** (0.058) in column (2); 0.1194** (0.055) in column (3).
    - Real GDP growth: 20.0056 (0.085) in (1); 0.1128 (0.076) in (2); 0.0518 (0.075) in (3).
    - Inflation (eop): 0.0166 (0.030) in (1); 20.0604 (0.053) in (2); 20.0420 (0.056) in (3).
    - NER (% change, eop): 20.0309* (0.016) in (1); 20.0073 (0.020) in (2); 20.0096 (0.022) in (3).
    - Public debt (% GDP): 0.0644** (0.030) in (1); 0.0773** (0.035) in (2); 0.0751** (0.031) in (3).
    - Credit/GDP: 20.0863*** (0.031) in (2).
    - Stock market capitalization (% GDP): 20.0319*** (0.011) in (2).
    - Number of listed companies (per 1,000,000 people): 20.0291* (0.015) in (3).
  - Model fit and coverage:
    - Observations: 1,763 in (1); 954 in (2); 1,002 in (3).
    - R-squared: 0.723 in (1); 0.789 in (2); 0.775 in (3).
    - Number of countries: 94 in (1); 70 in (2); 70 in (3).
  - Interpretation: results support the liquidity view — banks hold more government debt during periods of high interest rates and in countries with lower private sector credit to GDP ratios; less developed financial systems correlate with higher government debt holdings.

### Behavior during sovereign distress and the euro area case
- Around sovereign crises (53 advanced and emerging economies, 1970–2014):
  - On average, banks do not change their exposure during periods of sovereign distress (Figure 1 reported in source).
  - Interpretation ambiguous: maintenance of exposure could reflect risk taking, financial repression, or banks being stuck with illiquid securities that are costly to sell.
- Euro area evidence (late 2007 to mid-2013):
  - ECB and Bankscope series show a steady increase in banks’ holdings of sovereign debt from approximately 4 percent to 6.5 percent of assets (ECB: Sep-07 total 4.2 → Jun-13 total 6.5; Bankscope: 2007–2013 mean total 5.1).
  - Increase driven almost exclusively by domestic debt; holdings of foreign sovereign debt remained ~1.5 percent of assets.
  - Domestic exposure evolution in distressed vs. non-distressed countries:
    - Distressed countries (Greece, Ireland, Italy, Portugal, Spain): domestic exposure increased from 2.5 percent to 7 percent of assets.
    - Non-distressed countries (Austria, Belgium, Finland, France, Germany, Luxembourg, the Netherlands): smaller increases reported.
  - Interpretation: the disproportionate rise in domestic holdings in distressed countries is consistent with risk-taking and financial repression explanations; limited increase in foreign sovereign holdings suggests liquidity alone may not explain the behavior.
- Additional test using exogenous variation in rollover needs (driven by stock of maturing sovereign debt) can help isolate domestic bank behavior versus other factors.

_Italic: Source — 45133-dp1816-managing-the-sovereign-bank-nexus - Section IV discusses this channel._

### 1. Domestic versus Foreign Exposure

### 1. Domestic versus Foreign Exposure

### Axis values (percent scale)
- 0%
- 1%
- 2%
- 3%
- 4%
- 5%
- 7%
- 6%
- 8%

### Timeline (dates shown)
- Sep-2007
- Dec-07
- Mar-08
- Jun-08
- Sep-08
- Dec-08
- Mar-09
- Jun-09
- Sep-09
- Dec-09
- Mar-10
- Jun-10
- Sep-10
- Dec-10
- Mar-11
- Jun-11
- Sep-11
- Dec-11
- Mar-12
- Jun-12
- Sep-12
- Dec-12
- Mar-13
- Jun-13

*Source: 45133-dp1816-managing-the-sovereign-bank-nexus - 1. Domestic versus Foreign Exposure*

### 2. Domestic Exposure: Periphery versus Core Countries

### 2. Domestic Exposure: Periphery versus Core Countries

### Domestic versus foreign sovereign exposures (periphery and core)
- Sample comprises 247 banks from euro area countries. Periphery countries include Greece, Ireland, Italy, Portugal, and Spain. Core countries include Austria, Belgium, Finland, France, Germany, Luxembourg, and the Netherlands.
- Sovereign exposures are expressed in percent of total bank assets.
- Banks represent an important share of sovereign debt markets, holding on average 30 percent of outstanding public debt in advanced economies and 45 percent in emerging markets (Arslanalp and Tsuda 2014).

### Moral suasion versus risk shifting — empirical evidence (Table 5)
- Sample: 76 euro area banks in the regressions reported.
- Finding: Domestic banks are considerably more likely to purchase domestic sovereign bonds than foreign banks in months when the need of financing by the domestic sovereign is high (albeit with a relatively low level of statistical significance).
- Evidence that banks with closer links to governments (direct ownership or board seats) increased exposure to domestic government debt disproportionately between 2010 and 2013, but only during periods of high sovereign risk (sovereign CDS spread exceeded 100 basis points) (Becker and Ivashina 2018; Brutti and Sauré 2015).
- Mixed evidence on risk shifting:
  - Acharya and Steffen (2015) find institutions with lower capital and riskier portfolios loaded up more on government debt during 2010–12.
  - In the sample in this chapter there is not a negative correlation between a bank’s holdings of domestic sovereign bonds and its net worth.
- Table 5 regression highlights (coefficients shown as in source):
  - High need * Domestic bank: 0.034* (0.019) in column (1); 0.035* (0.021) in column (2).
  - Size * Domestic bank: 0.006 (0.109); 0.001 (0.136).
  - Capital * Domestic bank: 0.487 (0.357); 0.873* (0.441).
  - Loans/Deposits: 0.062* (0.038); 0.059 (0.040).
  - Bank FE, Country-year-month FE: Yes in reported specifications.
  - Number of banks: 41 in Greece, Ireland, Italy, Portugal, and Spain (column 1) and 36 in Ireland, Italy, Portugal, and Spain (column 2).
  - Sample periods: May 2010–August 2012 for banks in Greece, Ireland, and Portugal; August 2011–August 2012 for banks in Italy and Spain.
  - “High need” = indicator for months when new debt auctioned by the domestic government is above the median.

### The Sovereign Exposure Channel — bank losses and market effects
- Example sensitivity: a 10 percent loss on a sovereign bond portfolio representing 10 percent of bank assets would imply a 15 percent reduction in bank capital for a bank with a 6.6 percent leverage ratio.
- Sovereign restructuring haircuts and estimated capital losses (Table 6; calculations based on 2005–13 bank balance sheet data):
  - [1] Sovereign exposure/Total assets: Means — All 9.57, AEs 7.7, EMDEs 15.9; Medians — All 6.5, AEs 5.4, EMDEs 14.0.
  - [2] Comm equity/Total assets: Means — All 7.1, AEs 6.4, EMDEs 10.6; Medians — All 6.1, AEs 5.8, EMDEs 9.3.
  - [3] Loss on sov exp after haircut: 37% (of total assets): Means — All 3.5, AEs 2.9, EMDEs 5.9; Medians — All 2.4, AEs 2.0, EMDEs 5.2.
  - [4] Loss on sov exp after haircut: 50% (of total assets): Means — All 4.7, AEs 3.9, EMDEs 8.0; Medians — All 3.2, AEs 2.7, EMDEs 7.0.
  - [5] Comm equity/Total assets after haircut: 37%: Means — All 3.6, AEs 3.5, EMDEs 4.7; Medians — All 3.7, AEs 3.8, EMDEs 4.2.
  - [6] Comm equity/Total assets after haircut: 50%: Means — All 2.3, AEs 2.5, EMDEs 2.6; Medians — All 2.9, AEs 3.1, EMDEs 2.3.
  - [7] Loss on sov exp after haircut: 37% (of common equity): Means — All 49.5, AEs 44.8, EMDEs 55.8; Medians — All 39.2, AEs 34.2, EMDEs 55.4.
  - [8] Loss on sov exp after haircut: 50% (of common equity): Means — All 66.9, AEs 60.5, EMDEs 75.5; Medians — All 53.0, AEs 46.3, EMDEs 74.9.
- Market event evidence (Table 7, EBA stress test releases):
  - Interaction coefficient Sovereign CDS * Domestic sovereign exposure on banks’ stock returns: −1.664*** (0.396) in column (1); similar magnitude and significance across columns.
  - Domestic sovereign exposure (% assets) coefficients and Sovereign CDS coefficients reported as in source; in stock-return regressions Sovereign CDS has large negative coefficients (e.g., −0.177*** (0.035)).
  - In CDS-spread regressions (columns 4–5), interaction terms imply a bank with 10 percent of domestic sovereign bond holdings has a CDS spread higher by 44.3 basis points than a bank with zero such holdings (column 4).
- Consequences documented:
  - Banks with greater exposures to risky sovereign debt experienced larger falls in stock market values once exposures were disclosed.
  - Banks in stressed countries have higher CDS spreads if they hold a larger share of assets in domestic sovereign debt; correlation of bank and sovereign CDS spreads is higher for such banks.
  - Banks with substantial holdings of distressed sovereign debt reduced syndicated lending by 21.3 percent relative to banks with marginal holdings as spreads rose toward end-2010 (Popov and van Horen 2015).

### The Safety Net Channel — guarantees, deposit rates, and fiscal costs
- Banks hold government backstops and guarantees—explicit (insured deposits, central bank facilities) and implicit (too-big-to-fail expectations).
- Sovereign distress can weaken credibility of backstops, increasing banks’ funding costs and reducing ability to attract funding; conversely, bank failures can impose fiscal burdens via recapitalization and lost dividend revenues.
- Evidence on deposit rates and sovereign CDS (Figure 3 and Table 8):
  - Before the euro-area crisis, sovereign CDS spreads were low and uncorrelated with bank deposit rates; during the crisis a strong positive correlation appeared.
  - Table 8 regression highlights (sample period 2006–13, half-year frequency, eight euro area countries):
    - Sovereign CDS coefficients across specifications: 0.5408*** (0.119) in column (1); 0.5033*** (0.141); 0.4892*** (0.129); 0.4755*** (0.119).
    - Bank CDS: 0.0777*** (0.020) in column (1); 0.0700*** (0.017) in column (2); −0.0238 (0.061) in column (3); −0.0070 (0.069) in column (4).
    - Interaction terms involving Bank CDS * Sovereign CDS and period dummies show varying significance; Bank CDS*Sovereign CDS*2006–07 = 0.0875** (0.036) in column (2).
  - Interpretation: sovereign CDS is a significant determinant of the deposit rate paid by banks, even after controlling for bank CDS; relationship stronger for weaker banks and weaker sovereigns.
- Event-study evidence on safety net (Table 9, EBA stress test releases):
  - Sovereign CDS * Domestic sovereign exposure: −1.7087*** (0.546) in column (1); similar in columns (2)–(3).
  - Sovereign CDS * Capital: 0.0727*** (0.020); Sovereign CDS * Risk profile: −0.0058*** (0.002).
  - Interpretation: When sovereign risk increases, fall in bank stock market value is greater for banks that (1) are highly exposed to government debt and (2) have weak balance sheets (lower capital ratios, higher asset riskiness), consistent with safety net channel.
- Fiscal costs of safety-net interventions and banking crises:
  - Direct fiscal outlays in systemic banking crises (1970–2011) — medians (Table 10):
    - Output loss: All 23.2 percent of GDP; Advanced 32.4; Emerging 33.6; Developing 0.7.
    - Increase in public debt: All 12.1 percent of GDP; Advanced 23.6; Emerging 9.1; Developing 10.9.
    - Direct fiscal costs: All 6.8 percent of GDP and 12.5 percent of financial system assets; Advanced 4.2 percent of GDP and 2.1 percent of financial system assets; Emerging 8.3 percent of GDP and 21.3 percent of financial system assets; Developing 10.0 percent of GDP and 18.3 percent of financial system assets.
  - Direct fiscal outlays can be very large in extremes (e.g., Indonesia >50 percent of GDP in 1997–2001 episode; Iceland and Ireland >30 percent of GDP in recent crisis).
  - Tradeoffs: guarantees (low initial disbursements but contingent liabilities) versus direct recapitalizations (immediate fiscal cost but partial recovery on divestment); timing matters—faster resolution often lowers ex-post fiscal costs.
  - Moral hazard risk: implicit and explicit guarantees can weaken ex ante monitoring and foster correlated risk-taking by banks.

### The Macroeconomic Channel — sovereign risk, fiscal consolidation, and growth
- When sovereign risk rises, governments often pursue fiscal consolidation (raising taxes or cutting spending), which can contract economic activity and worsen firm revenues and household income.
- Higher sovereign spreads also raise corporate borrowing costs and affect household wealth—these effects increase nonperforming loans and hinder bank profitability.
- Evidence on banking crises and macro performance:
  - Banking crises disrupt credit supply, leading to deeper and longer recessions; average cumulative output losses associated with banking crises estimated at about 23 percent of GDP (Laeven and Valencia 2013a).
  - Recessions preceded by rapid banking expansions tend to be deeper, longer, and have slower recoveries:
    - Banking expansion defined as five-year cumulative change in private-sector credit-to-GDP above median; "severe" if in top quartile.
    - Median recovery time for recessions preceded by severe banking expansions is more than twice that for recessions preceded by nonsevere expansions.
  - Fiscal dynamics: primary balances tend to be higher during banking expansions but deteriorate more quickly during ensuing downturns; fiscal sustainability pressures last longer in banking recessions.
  - Cross-border transmission through banks’ subsidiary networks can transmit real effects internationally.

### Discussion and policy implications
- Four main findings summarized:
  - The sovereign-bank nexus operates through multiple interacting channels: sovereign exposure channel, safety net channel, and macroeconomic channel—all empirically relevant.
  - Banks hold substantial government debt, leaving them vulnerable to sovereign distress; exposure explains impact on bank stability. Banks use public debt for liquidity management and broader financial support; some weak evidence for risk-taking and moral suasion motives in the euro area.
  - Clear link between the strength of a country’s safety net and bank financial stability: interest rates on deposits depend on sovereign CDS spreads; relationship stronger for precarious banks that rely more on safety net.
  - Banking crises trigger implicit and explicit guarantees and ad hoc interventions with significant fiscal impacts; direct fiscal costs of banking crises between 1970 and 2011 averaged as reported above.
- Policy-relevant considerations (implied by empirical findings in chapter):
  - Monitor and limit excessive domestic sovereign concentration in bank portfolios to reduce vulnerability to sovereign distress.
  - Strengthen resolution frameworks and implement measures to reduce implicit guarantees (e.g., higher capital for systemic institutions, recovery and resolution planning, TLAC/bail-inable debt) to mitigate safety net channel; note these reforms are largely still in implementation and not reflected in empirical analysis.
  - Weigh timing and form of interventions—guarantees versus recapitalizations—considering fiscal capacity and tradeoffs between direct fiscal outlays and indirect macroeconomic costs.
  - Recognize interaction of sovereign and bank risks for macroprudential and fiscal policymaking: sovereign stress can amplify bank fragility and vice versa, so coordinated frameworks are important.

*Italic: Source: Managing the Sovereign-Bank Nexus — Chapter 2 (Domestic Exposure: Periphery versus Core Countries), IMF DP content provided.*

### 6.8 percent of GDP, with larger numbers in emerging markets and devel-

### Managing the Sovereign-Bank Nexus

### Key findings
- Banking crises have an exogenous negative impact on economic activity and additional adverse effects on fiscal accounts.
- Pressures on fiscal balances last longer in banking recessions (recessions associated with problems originating in the banking sector) than in standard recessions.
- Banking crises act as a multiplier of other negative shocks; associated increases in public debt are far greater (at least in advanced economies) than those ascribed to direct financial sector support.
- On average, banking crises are associated with a cumulative output loss of 23 percent of GDP.
- Public debt increases in the wake of banking crises by about 12 percent of GDP.
- Sovereign-bank bilateral feedback can generate multiplier effects that increase with the extent of debt held by banks; evidence of such “doom loops” was observed in the euro area sovereign-debt crisis where sovereign and bank credit risk became more correlated following bank bailout announcements.
- 6.8 percent of GDP, with larger numbers in emerging markets and developing economies. (As presented in the source text.)

### Policy implications (general)
- Strengthening both bank and sovereign balance sheets and governance is essential to reduce the likelihood and potency of the sovereign-bank nexus.
- Larger fiscal buffers and better public-debt management reduce the probability of sovereign-related bank distress.
- Larger capital buffers and improved regulation and supervision reduce bank risk taking and boost banks’ ability to withstand shocks.
- Countries with large banking systems and strong bank-to-sovereign linkages may justify larger bank capital buffers and stronger fiscal buffers.
- Policies must be designed and evaluated holistically because interventions on one channel can produce unintended effects on others.
- Post-crisis regulatory reforms have improved crisis management tools, resolution frameworks, and banks’ loss-absorption capacity, but more is needed to further weaken the nexus.

### Sovereign exposure channel — findings and recommended measures
- Problem: Sovereign exposures are often given favorable regulatory treatment (zero risk weight for domestic sovereign exposures denominated and funded in domestic currency), incentivizing excessive holdings and domestic home bias.
- Risks from full alignment with risk: Establishing positive, risk-sensitive capital requirements and strict concentration limits would eliminate the regulatory incentive to hold sovereign paper but may make the system more procyclical.
- Tradeoffs:
  - Risk-sensitive risk weights increase capital requirements during sovereign distress, encouraging banks to reduce exposures, potentially weakening the debt-stabilizing behavior of banks and strengthening the macroeconomic channel.
  - Interaction with liquidity requirements could force reductions in other credit exposures, weakening the sovereign exposure channel but strengthening the macroeconomic channel.
- Preferred regulatory approaches:
  - Impose a positive but time-invariant (through-the-cycle) risk weight for domestic sovereign exposures to reduce the incentive to invest excessively in domestic sovereign bonds while avoiding regulatory-induced procyclicality.
  - Calibrate capital requirements that increase with sovereign concentration on banks’ balance sheets.
  - Consider capital surcharges reflecting concentrations of sovereign holdings to discourage home bias and incentivize a more diverse pool of liquid assets.
  - Consider replacing domestic sovereign exposures with securities backed by a diversified pool of assets across countries (sovereign-bond-backed securities) to spread default risk and reduce idiosyncratic sovereign exposure.
- Additional observations:
  - Exposures to central banks may merit more lenient treatment than exposures to central governments, subject to strong governance arrangements for monetary policy implementation.
  - The regulatory definition of sovereign exposures is broad and often includes loans to subnational governments and public sector entities, which do not play the same liquidity and collateral role as central government bonds; such instruments do not justify the same favorable regulatory treatment.
  - Additional transparency and disclosure (breaking down sovereign exposures by jurisdiction, currency denomination, and account classification as well as their regulatory treatment) could foster market discipline and reduce excessive sovereign holdings.

### Safety net channel — findings and recommended measures
- Objective: Weaken the safety net channel by restricting and reducing the need for government guarantees on banks and introducing private loss-sharing arrangements in bank resolution.
- Progress since the financial crisis:
  - Systemically important banks are now subject to tighter loss absorbency and supervisory requirements.
  - Introduction and upgrading of resolution regimes provide authorities with a broader range of resolution tools to enable orderly resolution of banks.
  - The new minimum requirement for liabilities (or loss-absorbing capacity) that should be readily available for bail-in within resolution is meant to support orderly resolution of global systemically important banks without recourse to public funds by making bail-in resolution strategies feasible and credible.
  - The standard is known as total loss-absorbing capacity (TLAC). It was agreed on by the Financial Stability Board in November 2015 and applies to global systemically important banks.
- Ex-ante policies: Bail-in policies that impose losses on shareholders and uninsured debtholders to minimize moral hazard and enhance market discipline.
- Ex-post policies: Improved resolution frameworks limit recourse to public funds and, if implemented effectively, weaken the macroeconomic channel by facilitating speedy resolution, preserving bank value, and minimizing spillovers.
- Caveat: Resolution regimes still need constrained flexibility to allow temporary public sector support in times of severe contagion risk to mitigate system-wide distress and ensure continuity of credit provision.
- Implementation priority: Fully and timely implement reforms focusing on robust early intervention powers, effective resolution toolkits and adequate loss-absorbing capacity, effective arrangements for cross-border cooperation, and credible recovery and resolution plans.

### Cross-border integration — findings and policy considerations
- Large internationally active banks add an international dimension to the nexus; cross-border operations can both weaken and strengthen different channels:
  - Diversified balance sheets across countries can weaken domestic macro and sovereign exposure channels.
  - However, internationally active banks may become larger and thereby strengthen the safety net channel because resolution remains national in practice (“global banks remain national in death”).
- Practical developments:
  - Jurisdictions are introducing intermediate holding company requirements that mandate local subsidiaries to operate and be resolvable independently, facilitating resolution and liquidity support but potentially reinforcing links between subsidiaries and their sovereign hosts and fragmenting cross-border banking flows.
- If full integration (for example, a full banking union) were achievable, the domestic nexus could be substantially weakened through:
  - An integrated safety net with centralized deposit insurance and lender-of-last-resort functions.
  - Centralized supervision and resolution frameworks that eliminate the need for banks to hold bonds from specific countries for liquidity management and foreign exchange risk mitigation, allowing prudential rules to limit concentration to individual sovereigns without compromising the bond-holding liquidity function.
  - Measures to facilitate cross-border banking integration, including physical presence abroad, that would help weaken the macro channel while an integrated safety net applies.

### Regulatory treatment summary (as presented)
- Under the Basel standardized approach, sovereign exposures are risk-weighted according to external ratings on an increasing scale; at national discretion a lower weight may be applied to domestic sovereign debt provided it is denominated and funded in domestic currency.
- All Basel Committee member jurisdictions apply a zero risk weight to domestic sovereign debt in domestic currency; this exemption is extended across the European Union.
- Under the internal-ratings-based approach, banks may use internal models to estimate default probabilities and loss-given-default; sovereign debt is exempt from the 3-basis-point floor under the default probability prescribed for private issuers with broadly similar characteristics.
- Risk Weights for Sovereign Exposures in the Standardized Approach (Percent):
  - Credit rating: AAA to AA– | Risk weight: 0
  - Credit rating: A+ to A– | Risk weight: 20
  - Credit rating: BBB+ to BBB– | Risk weight: 50
  - Credit rating: BB+ to B– | Risk weight: 100
  - Credit rating: Below B– | Risk weight: 150
  - Credit rating: Unrated | Risk weight: 100

*Source: Managing the Sovereign-Bank Nexus (excerpt).*

### Box 1. regulatory treatment of Sovereign Exposures

### Box 1. regulatory treatment of Sovereign Exposures

### Risk weighting and capital treatment
- Under the Basel framework, the risk weights applied to domestic sovereign debt under the credit risk–based approach are often close to zero.
- Sovereign exposures denominated and funded in domestic currency can receive a preferential default risk charge under the standardized market risk approach (national discretion).
- Under the internal-ratings-based approach, sovereign exposures should be included in models.

### Supervisory actions and stress testing
- Some jurisdictions (such as the EU) perform stress tests of banks that include shocks to sovereign exposures.
- Results from these stress tests may guide supervisory actions concerning additional (Pillar 2) capital requirements.
- Following consultation, the Committee decided in 2016 to enhance the Pillar 2 approach for interest rate risk in the banking book.

### Credit risk mitigation and repos
- The Basel framework permits a national discretion to apply a zero haircut for repo-style transactions where the collateral is a sovereign security with core market participants.

### Interest rate risk
- Sovereign debt is exposed to significant interest rate risk.
- There is no Pillar 1 charge for interest rate risk; the response has been to enhance the Pillar 2 approach (decision in 2016).

### Large exposures and concentration limits
- There is currently no large exposure or concentration limit applied to sovereign exposure.
- Sovereign debt is explicitly exempt from the large exposure requirement that limits exposures to any single counterparty or group of connected counterparties to 25 percent of eligible capital.

### Leverage ratio
- Sovereign exposures are included in total assets in the leverage ratio calculations.
- The leverage ratio is not a binding constraint for most banks and, thus, does not de facto restrict sovereign exposures.

### Liquidity standards (Liquidity Coverage Ratio)
- Under the liquidity coverage ratio, no limits or haircuts are applied to sovereign exposures.
- Sovereign exposures are defined as high-quality liquid assets (those assigned a zero percent risk weight under the standardized approach).

*Box 1. regulatory treatment of Sovereign Exposures — Managing the Sovereign-Bank Nexus (excerpt).*

---


_Source: https://www.imf.org/-/media/files/publications/dp/2018/45133-dp1816-managing-the-sovereign-bank-nexus.pdf_