## Executive Summary — IMF Financial Sector Assessment Program (FSAP) — Denmark (1dnkea2020001)

## Source details

**Canonical URL:** [Executive Summary — IMF Financial Sector Assessment Program (FSAP) — Denmark (1dnkea2020001)](https://www.imf.org/-/media/files/publications/cr/2020/english/1dnkea2020001.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2020/english/1dnkea2020001.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2020/english/1dnkea2020001.pdf.json)

---

### Background and macrofinancial context
- Much of the FSAP work was conducted prior to the COVID-19 pandemic; findings and recommendations remain pertinent for medium-term challenges and vulnerabilities.
- Report updated to reflect key developments and policy changes since the FSAP mission and includes illustrative scenarios to quantify possible implications of the COVID-19 shock on the solvency of systemically important financial institutions (SIFIs).
- COVID-19 illustrative scenarios focus on growth and unemployment implications; they do not entail an abrupt tightening of financial conditions or a real estate downturn except in a third scenario with more severe market shocks but less severe economic shocks. Amplification effects on banking liquidity or potential system-wide contagion risks were not assessed in these scenarios.
- Authorities provided discretionary fiscal support of approximately DKK 126 billion (5½ percent of 2019 GDP). Danmarks Nationalbank (DN) introduced an extraordinary lending facility and reestablished swap lines; the countercyclical capital buffer (CCyB) was released.
- The 2014 FSAP found broadly contained risks but called for additional measures; some weaknesses identified then remain, and the Danske Bank money laundering case highlighted longstanding supervisory deficiencies.

### Financial system size, structure, and key metrics
- Financial system assets: over 630 percent of GDP.
- Banking sector: 77 banks and mortgage credit institutions (MCIs); accounts for 55 percent of financial sector assets.
- Dominance: seven domestic SIFIs; Danske Bank Group assets about 170 percent of GDP.
- Insurance companies and pension funds (ICPFs): 190 percent of GDP.
- ATP: dominant provider of pension savings outside the insurance sector.
- Covered bond market: about 130 percent of GDP (almost five times the size of the government debt market).
- Aggregate risk-weighted capital ratio: 22.5 percent at end-2019.
- Aggregate CET1 ratio: 18.3 percent in 2020 Q1 (an increase of over a full percentage point before the crisis).
- All SIFIs met minimum leverage ratio requirement: 3 percent.
- Liquidity Coverage Ratio (LCR): all banks exceeded minimum 100 percent; median LCR across systemic banks was 198 percent as of April 2020.
- Nonperforming loan (NPL) ratio: 1.7 percent at end-2019.
- Bank profitability: aggregate ROE about 11½ percent on average over the last four years prior to COVID-19; ROE in 2020 Q1 fell to -2.7 percent.
- Investment funds: account for about 91 percent of GDP.

### Macrofinancial vulnerabilities (on eve of COVID-19)
- Household debt: high level of household (gross) debt; household net wealth relatively high but assets largely illiquid (housing and pension savings).
- Real estate: nationwide house prices stabilized; urban house prices remained elevated through 2019; downside risks larger in Copenhagen.
- Mortgage loans: variable-rate mortgages constitute 60 percent of the stock; 36 percent of mortgages are non-amortizing. (More than 50 percent of variable-rate mortgages have fixed-interest-rate periods greater than one year.)
- CRE: accounts for about 30 percent of total corporate debt; six-fold jump in debt-at-risk in an adverse scenario pre-pandemic.
- Covered bond exposures account for 60 percent of all fixed-income assets and are almost entirely domestic for ICPFs.

### COVID-19 illustrative scenarios (five-year, not forecasts)
- "COVID central":
  - GDP: sharp fall of 8 percent in 2020, rebound of 6 percent in 2021.
  - Unemployment: rises to 6.8 percent in 2020, then gradually declines.
  - Assumes no abrupt tightening of financial conditions or real estate downturn.
- "COVID prolonged":
  - GDP: decline of 8 percent in 2020, further contraction of 2.1 percent in 2021, recovery in 2022.
  - Unemployment: rises to 11.8 percent in 2022 before steadily falling.
  - Assumes no sharp tightening in financial conditions or decline in house prices.
- "Market shocks":
  - Sharp rise in asset risk premia, adverse effects on banks’ funding costs, and a prolonged domestic real estate downturn amid a protracted recession less severe in the short term relative to the other scenarios.
  - Incorporates credit, market, and cost-of-funding shocks; designed before COVID-19.

### Solvency stress-test outcomes (seven SIFIs; consolidated group-level)
- COVID central scenario:
  - Aggregate CET1 ratio declines from 18.3 percent in 2019 to a low point of 13.4 percent in 2020.
  - Aggregate leverage ratio declines to 3.6 percent.
  - All SIFIs meet minimum capital requirements (4.5 percent CET1).
  - One of seven SIFIs must partially tap its capital conservation buffer (CCB).
  - Leverage ratio for one SIFI falls below three percent (leverage ratio planned for implementation in mid-2021).
- COVID prolonged scenario:
  - Aggregate CET1 ratio falls to a low point of 12.4 percent in 2022.
  - Aggregate leverage ratio declines to 3.0 in 2024.
  - All SIFIs still meet minimum capital requirements; one SIFI partially uses CCB; another partially uses SIFI buffer.
  - Leverage ratios of a few SIFIs fall below three percent.
- Market shocks scenario:
  - Aggregate CET1 ratio decreases to 13.4 percent in 2024 (trough at end of horizon).
  - Aggregate leverage decreases to 3.4 percent.
  - A few SIFIs partially use CCBs; a few SIFIs’ leverage ratios dip slightly below three percent.
- Interpretation and limitations:
  - Relationship between growth and capitalization is non-linear; deeper recessions can have disproportionate impacts.
  - Some second-round and non-linear effects not fully captured; possible underestimation of losses.
  - Further deterioration or tighter financial conditions could cause some SIFIs to breach minimum capital requirements.

### Liquidity stress tests — key findings
- Tests assumed shocks significantly larger than those observed in 2020 Q1.
- All five banks assessed would be able to survive extreme liquidity pressures without external support for longer than three months using cash and liquidating securities (including via central bank standard operations).
- Bank liquidity appears adequate at present, but tighter conditions relative to assumptions could cause acute liquidity strains for some banks.
- MCIs play a central role in domestic interbank system and can generate significant contagion effects in a system densely linked by covered bond exposures.

### Insurance sector stress tests — design and outcomes
- Coverage: five life insurers, six occupational pension funds, five non-life insurers; ATP included.
- Adverse market shocks were main risk factor—tests considerably more severe than 2020 Q1 market moves.
- Bottom-Up (BU) results:
  - Median SCR ratio in life sector fell from 188 to 127 percent after stress.
  - Median SCR ratio in non-life sector fell from 233 to 207 percent after stress.
  - One life company drops below solvency ratio of 100 percent.
  - Capital needed to restore full coverage would be a moderate fraction of the sample's aggregate capital.
- Top-Down (TD) results broadly confirm BU results.
- Medium-term implications:
  - Life insurers face declining investment returns as higher-coupon bonds expire.
  - In aggregate, Danish life insurance sector records positive spreads of investment returns over guaranteed interest rates and is expected to regain profitability after adverse scenario, with significant differences across companies.
  - Insurers with high legacy guarantees likely to experience a drain on profitability.
- Sensitivity tests:
  - Longevity shock: strong increase in life expectancy reduces median SCR of life insurers and occupational pension funds to 210 percent.
  - Catastrophic events: repetition of windstorm Anatol would reduce non-life insurers’ own funds by only one percent due to reinsurance.

### Interconnectedness and contagion analysis
- Multi-layer contagion model distinguishes transmission across eight exposure types using credit registry data.
- Key findings:
  - MCIs are central nodes inducing highest levels of contagion losses via unlisted shares (complex group structures) and covered bond exposures.
  - Contagion transmitted predominantly through credit losses.
  - Systemic banks on average more vulnerable to inward spillovers because of covered bond holdings.
  - Four financial institutions identified with the ability to amplify spillovers given high contagion and vulnerability scores.
  - Cross-border exposures strong with the Nordic region and the euro area.

### Macroprudential framework, institutional setup, and recommendations
- Institutional features:
  - Systemic Risk Council (SRC) chaired by chairman of Board of Governors of DN; SRC advisory only with comply-or-explain mechanism.
  - Ultimate decision power rests with minister of industry, business and financial affairs (MIBFA); hard powers effectively lie with the government.
  - SRC secretariat housed in DN; resource constraints limit publications and proactive drafting.
- Countercyclical capital buffer (CCyB):
  - CCyB increased to two percent (effective December 2020) and fully released following onset of COVID-19 pandemic.
- Sources of inaction bias:
  - SRC legal framework emphasizes striving for consensus and starts drafting only after mandate, creating delays and implicit veto power.
  - No legal basis to introduce binding borrower-based macroprudential measures; many measures implemented as guidelines under consumer protection legislation.
- Policy recommendations (selected, implementing authority and timing where provided):
  - Give SRC chair the ability, enshrined in law, to make proposals for a recommendation after due consultation without needing consensus (MIBFA). MT
  - Introduce national legislation to include borrower-based tools (limits on LTVs, DTIs, and DSTIs) in the policy toolkit (MIBFA). MT
  - Introduce stricter LTV limits to safeguard against large house price shocks (MIBFA). MT
  - Introduce a binding income-based macroprudential measure that limits lending to households above a certain DTI (MIBFA). MT
  - Issue recommendations to reduce debt bias, simplify rental regulations, and relax supply constraints on housing (SRC). MT
  - Close data gaps, including enhancing coverage and quality of CRE data (DN). MT
  - Develop macroprudential stress tests that incorporate feedback loops between the financial system and real economy and contagion effects (DN). MT

### Financial sector oversight — supervision, AML/CFT, and macroprudential policy
- Prudential supervision:
  - DFSA's risk-based approach emphasizes credit risk in banking and market risks and asset-liability management in insurance.
  - Supervision needs to strengthen governance and risk culture oversight, including non-financial risks (compliance, governance, conduct, compensation, risk culture).
  - Insurance supervision should increase on-site inspection frequency, finalize a solid risk assessment framework, and enhance cross-border oversight.
  - DFSA’s operational independence should be safeguarded and its resources increased.
- AML/CFT:
  - Recent legislation (entered into force January 2020) strengthens DFSA monitoring and enforcement powers, including administrative fine notices and stiffer penalties.
  - DFSA’s AML Division expanded since June 2019; new institutional risk assessment model developed but not finalized.
  - Recommendations:
    - Test, finalize, and implement DFSA’s institutional risk assessment model (DFSA). ST
    - Intensify AML/CFT on-site inspections of higher-risk financial institutions (DFSA). ST
    - Consider options for regional consolidation of AML/CFT supervision (MIBFA). MT

### Systemic liquidity, Danmarks Nationalbank actions, and recommendations
- DN actions during crisis:
  - Launched extraordinary liquidity facilities and reactivated bilateral swap lines with ECB and the Federal Reserve.
  - Standing swap line with ECB doubled to EUR 24 billion; agreement with Federal Reserve for USD 30 billion swap line for at least 6 months.
  - DN increased policy rate by 15 basis points to –0.6 percent; maintains krone’s tight peg to the euro.
- Recommendations for DN:
  - Continue to refine operational preparedness for nonstandard liquidity support.
  - Complete framework for accepting credit claims as non-standard collateral (DN). ST
  - Improve domestic interagency collaboration and further automate framework for accepting credit claims as nonstandard collateral (DN, DFSA). ST
  - Communicate specific information requirements for ELA to key financial institutions.

### Crisis management and financial safety nets
- Improvements since 2014:
  - New resolution and deposit insurance legislation; resolution framework for banks and MCIs; designation of DFSA and Financial Stability Company (FSC) as resolution authorities; governance changes to the deposit insurance system; revival of Nordic-Baltic cooperation; joint crisis simulation in 2019.
- Recommendations:
  - Expedite resolution planning for SIFIs and finalize priority areas for resolution planning (DFSA, FSC). ST
  - Define strategies for liquidity assistance to institutions in resolution; DN should be able to provide liquidity to enable resolution plans, subject to safeguards (DN, DFSA, FSC, MIBFA, MOF). ST
  - Strengthen FSC autonomy, limit government involvement in resolution decisions to fiscal-support cases, and ensure Board independence and tenure stability (MIBFA). ST
  - Develop and test a national crisis management plan, including crisis communication and system-wide contingency planning (DFSA, DN, FSC, MIBFA, MOF). ST

### Climate-related financial risks
- Physical risk — flood exposure:
  - Danmarks Nationalbank (2019) estimates currently "1½ percent" of mortgage assets serving as collateral are at risk of flooding; this share could increase to "13 percent".
  - Current assessments indicate "1-3 percent" of SIFIs’ mortgaged assets are at risk of flooding; for two SIFIs the estimate could exceed "15 percent".
- Transition risk:
  - Financial sector has relatively limited exposure to highest carbon-emitting industries.
  - Credit institutions’ lending to less carbon-intensive industries accounts for "23 percent" of corporate sector lending.
  - Transition risks larger for institutions with large exposures to the agricultural sector and the energy sector.

### Key recommendations (summary)
- Systemic Risk Oversight and Macroprudential Policy:
  - Give SRC chair ability to make proposals for recommendations without consensus (MIBFA). MT
  - Introduce national legislation for borrower-based tools (LTV, DTI, DSTI) (MIBFA). MT
  - Introduce stricter LTV limits (MIBFA). MT
  - Introduce binding income-based macroprudential measure limiting lending above a certain DTI (MIBFA). MT
  - Issue recommendations to reduce debt bias, simplify rental regulations, relax housing supply constraints (SRC). MT
  - Close CRE data gaps and develop macroprudential stress tests with feedback loops (DN). MT
- Banking and Insurance Supervision:
  - Enhance DFSA operational independence and clearly state dismissal reasons for Director General (MIBFA). MT
  - Expand DFSA budget and recruit staff focused on non-financial risks (MIBFA). ST
  - Develop detailed guidance on risk assessments and enhance verification of supervisory data (DFSA). ST/MT
  - Increase insurance on-site inspections and strengthen cross-border supervision (DFSA). ST
- AML/CFT:
  - Finalize and implement DFSA’s institutional risk assessment model and intensify inspections (DFSA). ST
  - Consider regional consolidation options for AML/CFT supervision (MIBFA). MT
- Systemic Liquidity:
  - Complete framework for accepting credit claims as non-standard collateral and seek greater interagency information sharing (DN, DFSA). ST
- Financial Crisis Management and Safety Nets:
  - Strengthen FSC autonomy and expedite resolvability of SIFIs (MIBFA, DFSA, FSC). ST
  - Define liquidity assistance strategies for institutions in resolution and develop system-wide contingency and crisis communication plan (DN, DFSA, FSC, MIBFA, MOF). ST

*IMF staff summary of chapter content (1dnkea2020001).*

### EXECUTIVE SUMMARY __________________________________________________________________________ 6

### EXECUTIVE SUMMARY

### Background and macrofinancial context
- Much of the FSAP work was conducted prior to the COVID-19 pandemic; findings and recommendations remain pertinent for medium-term challenges and vulnerabilities.  
- The report has been updated to reflect key developments and policy changes since the FSAP mission work was completed and includes illustrative scenarios to quantify possible implications of the COVID-19 shock on the solvency of systemically important financial institutions (SIFIs).  
- The COVID-19 illustrative scenarios focus on the growth and unemployment implications of the shock; they do not entail an abrupt tightening of financial conditions or a real estate downturn except in a third scenario with more severe market shocks but less severe economic shocks. Amplification effects on banking liquidity or potential system-wide contagion risks were not assessed in these scenarios.  
- Prior to COVID-19, authorities had actively used macroprudential tools, strengthened banking and insurance supervision, enhanced AML/CFT supervision through recent legislation, and implemented a new bank resolution framework improving the financial safety net and crisis management frameworks.  
- The authorities provided discretionary fiscal support of approximately DKK 126 billion (5½ percent of 2019 GDP). Danmarks Nationalbank (DN) introduced an extraordinary lending facility and reestablished swap lines; the countercyclical capital buffer (CCyB) was released.  
- The 2014 FSAP found broadly contained risks but called for additional measures; some weaknesses identified then remain, and the Danske Bank money laundering case highlighted longstanding supervisory deficiencies.

### Systemic risk analysis and stress-testing outcomes
- Stress tests using illustrative COVID-19 scenarios show a large impact on SIFIs' capitalization ratios, but:
  - All SIFIs would continue to meet minimum capital requirements under the scenarios.
  - The impact is heterogeneous across SIFIs; in some cases a few SIFIs would need to partially use their capital conservation and/or SIFI buffers.
  - Given uncertainty and downside risks, a further deterioration of economic prospects or tighter financial conditions could cause some SIFIs to breach minimum capital requirements.
- Insurance sector stress tests:
  - Life insurers would be significantly hit by market shocks but most remain well above regulatory solvency thresholds.
  - Occupational pension funds and non-life firms are more resilient.
- Bank liquidity:
  - At present, bank liquidity appears adequate.
  - Tests suggest banks can withstand more significant funding pressures than those observed in March 2020.
- Mortgage Credit Institutions (MCIs):
  - MCIs play a central role in the domestic interbank system and can generate significant contagion effects in a system densely linked by covered bond exposures.

### Financial sector oversight — supervision, AML/CFT, and macroprudential policy
- Prudential supervision is generally sound, focusing on traditional financial risks:
  - DFSA's risk-based supervisory approach emphasizes credit risk in banking and market risks and asset-liability management in insurance.
  - Supervision needs to strengthen governance and risk culture oversight, including greater attention to non-financial risks which can rapidly undermine banking soundness.
  - Insurance supervision should increase on-site inspection frequency, complete a solid risk assessment framework, and enhance oversight of cross-border business.
  - DFSA’s operational independence should be safeguarded and its resources increased.
- AML/CFT:
  - Recent legislation enables enhanced AML/CFT supervision, strengthening monitoring and enforcement powers and stiffening certain non-compliance penalties.
  - Recommendations include finalizing DFSA’s institutional risk assessment model, intensifying on-site inspections of higher-risk institutions, and considering regional/EU-level consolidation options for AML/CFT supervision.
- Macroprudential policy:
  - Institutional arrangements can be improved by streamlining decision-making to reduce inaction bias and enhancing accountability.
  - If elevated housing vulnerabilities persist after the pandemic, additional macroprudential measures could be warranted, including tighter loan-to-value limits and binding income-based measures.
  - Coordinated policies to reduce debt bias, simplify rental regulations, and relax housing supply constraints are recommended.
  - Data gaps in the commercial real estate (CRE) sector limit vulnerability assessment and should be closed; systemic risk monitoring should better cover the nonbank sector.

### Systemic liquidity and central bank actions
- Danmarks Nationalbank (DN) promptly provided liquidity support during the crisis:
  - Launched extraordinary liquidity facilities and reactivated bilateral swap lines with the ECB and the Federal Reserve.
- Recommendations for DN:
  - Continue to refine operational preparedness for nonstandard liquidity support.
  - Improve domestic interagency collaboration and further automate the framework for accepting credit claims as nonstandard collateral.

### Crisis management and financial safety nets
- The financial safety net and crisis management framework have been considerably enhanced, but operational readiness needs further improvement.
- Resolution and crisis-management recommendations:
  - Expedite the resolution planning process for SIFIs and finalize priority areas for resolution planning.
  - Define strategies for liquidity assistance to institutions in resolution.
  - Strengthen the autonomy of the resolution authority (FSC), limiting government involvement in resolution decisions to cases requiring fiscal support.
  - Develop and test a national crisis management plan, including a crisis communication strategy.

### Climate-related financial risks
- Banks are more exposed to physical climate risk, especially flood-related disasters, than to transition risk:
  - Currently 1½ percent of mortgage assets serving as collateral are at risk of flooding; this share could increase to 13 percent.
  - Banks have limited exposure to high carbon-emitting industries.
- Non-life insurers are mainly exposed to windstorms and cloudbursts; these risks are largely transferred to reinsurers.

### Key recommendations (summary from Table 1)
- Systemic Risk Oversight and Macroprudential Policy (implementing authority and timing shown where provided):
  - Give the chair of the SRC the ability, enshrined in law, to make proposals for a recommendation after due consultation without needing consensus (MIBFA). MT
  - Introduce national legislation to include borrower-based tools (limits on LTVs, DTIs, and DSTIs) in the policy toolkit (MIBFA). MT
  - Introduce stricter LTV limits to safeguard against large house price shocks (MIBFA). MT
  - Introduce a binding income-based macroprudential measure that limits lending to households above a certain DTI (MIBFA). MT
  - Issue recommendations to responsible authorities to reduce debt bias, simplify rental regulations, and relax supply constraints on housing (SRC). MT
  - Close data gaps, including enhancing coverage and quality of CRE data (DN). MT
  - Develop macroprudential stress tests that incorporate feedback loops between the financial system and real economy and contagion effects (DN). MT
- Banking and Insurance Supervision:
  - Enhance DFSA operational independence, for example by clearly stating in law the reasons for dismissal of a DFSA Director General (MIBFA). MT
  - Expand DFSA budget to recruit and retain staff with a focus on non-financial risks (MIBFA). ST
  - Develop more detailed guidance on risk assessments to support supervisory judgement (DFSA). ST
  - Enhance approaches to confirm veracity of supervisory data and information (DFSA). MT
  - Increase insurance on-site inspections, finalize risk assessment framework, and strengthen cross-border supervision (DFSA). ST
- AML/CFT:
  - Test, finalize, and implement DFSA’s new institutional risk assessment model (DFSA). ST
  - Intensify AML/CFT on-site inspections of higher-risk financial institutions (DFSA). ST
  - Consider and pursue options for regional consolidation of AML/CFT supervision (MIBFA). MT
- Systemic Liquidity:
  - Complete framework for accepting credit claims as non-standard collateral (DN). ST
  - Seek greater domestic interagency information sharing for non-standard liquidity support preparedness (DN, DFSA). ST
- Financial Crisis Management and Safety Nets:
  - Strengthen FSC autonomy, limiting MIBFA decision power in resolution to fiscal-support cases (MIBFA). ST
  - Expedite resolvability of SIFIs, finalize priority areas for resolution planning (DFSA, FSC). ST
  - Define strategies for liquidity assistance to institutions in resolution (DN, DFSA, FSC, MIBFA, MOF). ST
  - Develop and test a system-wide contingency and crisis communication plan (DFSA, DN, FSC, MIBFA, MOF). ST

*Executive Summary — IMF Financial Sector Assessment Program (FSAP) — Denmark*

### 5.      Denmark’s financial system is large, with assets over 630 percent of GDP (Table 4). The

### 5.      Denmark’s financial system is large, with assets over 630 percent of GDP

### Financial system size and composition
- Financial system assets: over 630 percent of GDP.
- Banking sector: 77 banks and mortgage credit institutions (MCIs); accounts for 55 percent of financial sector assets.
- Dominance: seven domestic SIFIs; Danske Bank Group assets about 170 percent of GDP.
- Insurance companies and pension funds (ICPFs): 190 percent of GDP.
- ATP: dominant provider of pension savings outside the insurance sector.

### Interconnectedness, households, and mortgage funding
- High interconnectedness among financial institutions, households, and corporates.
- Household asset/liability structure: large shares of GDP in housing, pension savings, and mortgages.
- MCIs fund household mortgages by issuing covered bonds held by banks, ICPFs, and increasingly foreign institutional investors.
- Covered bond market: about 130 percent of GDP (almost five times the size of the government debt market); characterized as the world’s largest in relative and absolute terms.
- ICPFs hold covered bonds as assets against insurance and pension liabilities to households.

### Group structure and business models of major SIFIs
- Major SIFIs typically include a commercial bank and an MCI.
- Commercial banks: financed with deposits, provide loans to non-financial firms, act as the customer-facing contract point.
- MCIs: capitalized vehicles that issue covered bonds to finance mortgage portfolios originated by the commercial bank.
- Differentiated models reflected in asset and liability compositions across banks and MCIs.

### Nordic linkages and cross-border operations
- Strong banking linkages with other Nordic countries concentrated in the Nordic-Baltic region.
- Mortgage lending in foreign countries often conducted via MCI subsidiaries to comply with local covered bond regulations; other foreign operations via branches.
- Nordea (Finland-headquartered) operates a large branch in Denmark; other Nordic banks have small domestic presences.
- Danske Bank: significant operations across the Nordic region via MCI subsidiaries and commercial banking branches.
- Exposure to the Baltic region: now negligible following Danske’s retrenchment after the ML case.

### Capitalization, liquidity, asset quality, and profitability (pre- and early-COVID)
- Aggregate risk-weighted capital ratio: 22.5 percent at end-2019.
- All SIFIs met minimum leverage ratio requirement: 3 percent.
- SIFIs’ aggregate CET1 ratio: increased by over a full percentage point to 18.3 percent in 2020 Q1 (six months prior to the crisis).
- SIFI actions: many announced dividend cancellations and share buyback cancellations to bolster capital.
- Risk-weighted asset floors (Basel III finalized) may lower regulatory capitalization ratios for four Danish banks using IRB models.
- Liquidity Coverage Ratio (LCR): all banks exceeded minimum 100 percent; median LCR across systemic banks was 198 percent as of April 2020.
- Nonperforming loan (NPL) ratio: 1.7 percent at end-2019, but risk of deterioration due to COVID-19.
- Bank profitability: aggregate ROE about 11½ percent on average over the last four years prior to COVID-19; ROE in 2020 Q1 fell to -2.7 percent due to a sharp rise in loan impairment charges.
- Lending by banks and MCIs to households and nonfinancial corporates has declined.

### Deposit pricing and interest rate environment
- To mitigate net interest income erosion, banks offering negative rates to depositors.
- Share of corporate deposits earning negative interest rates in systemic banks: rose from about 40 to 60 percent over 2017-19.
- Many SIFIs passing on negative policy rates to households with deposits over DKK 750,000 (€100,000) — equivalent to current level of deposit insurance.
- Limitations/risks: cannot fully offset decline in net interest income if rates remain persistently low and could destabilize funding base.

### Insurance sector and investment funds
- Life insurance companies: challenged by prolonged low interest rates; Q1 2020 brought substantial investment losses.
- Legacy contracts: guaranteed rates of return of more than three percent (declining but large) which are above market rates.
- Product shift: market risks increasingly borne by policyholders via unit-linked contracts.
- Covered bond exposures: account for 60 percent of all fixed-income assets and are almost entirely domestic.
- Non-life insurance: favorable underwriting results and relatively low expense ratios.
- Solvency ratios: broadly stable, hovering around 280 percent in both life and non-life segments (regulatory threshold 100 percent).
- Investment funds: account for about 91 percent of GDP; expanding due to net inflows and rising valuations; have sizeable linkages with ICPFs, banks, and nonfinancial firms.
- Need: more granular breakdown and linkage analysis of investment funds for monitoring risks.

### Macrofinancial vulnerabilities (on eve of COVID-19)
- Household debt: high level of household (gross) debt remains an important vulnerability despite a gradual downtrend; household net wealth relatively high but assets largely illiquid (housing and pension savings).
- Real estate prices: nationwide house prices stabilized; urban house prices remained elevated through 2019; downside risks larger in Copenhagen relative to national market.
- Mortgage loans: variable-rate mortgages constitute 60 percent of the stock; 36 percent of mortgages are non-amortizing. (Note: more than 50 percent of variable-rate mortgages have fixed-interest-rate periods greater than one year.)
- Greater risk taking: evidence across banking and insurance sectors—medium-sized banks extending credit to households with high loan-to-income ratios; insurers investing in riskier assets and shifting to unit-linked products.
- NFC sector and commercial real estate (CRE): rising leverage in cyclically-oriented firms; loosening credit standards for NFC sector; CRE characterized by rising prices, large share of foreign investors, concerns about income-generating capacity; CRE accounts for about 30 percent of total corporate debt and showed a six-fold jump in debt-at-risk in an adverse scenario pre-pandemic.
- Covered bond market: strong reliance on market-based funding; increasing share of foreign investors diversifies but increases sensitivity to global investor sentiment; substantial domestic holdings across institutions.
- AML/CFT: incomplete implementation of risk-based AML/CFT supervision and a recent money laundering case that affected reputation and confidence; bond spreads and equity prices were adversely affected by the Danske Bank case in 2019, elevating Danske’s funding costs and to a lesser extent those of other institutions.

### Risks (Risk Assessment Matrix summary)
- Prolonged COVID-19 outbreak: containment measures could remain or be reintroduced through early 2021.
- Macroeconomic transmission: prolonged containment and uncertainty could reduce supply and domestic/external demand, weaken exports, investment, consumption; exacerbate corporate losses, unemployment, NPLs, loan loss impairments, and banking profitability; possible tighter credit conditions and adverse Nordic spillovers.
- Financial tightening and asset price declines: deteriorating fundamentals and lower risk appetite could trigger a second wave of financial tightening, protracted asset price declines affecting ICPFs, household wealth, consumption, and regional spillovers due to Nordic integration.
- Sharp house price correction: could be triggered by global, regional, or domestic shocks; close financial linkages could spark rapid regional spillovers and amplify impacts through tighter financial conditions and depressed activity.

### Stress testing approach and scenarios
- Coverage: stress tests of banks, MCIs, and insurance companies covering all seven SIFIs (representing 77 percent of assets of credit institutions). Tests performed at consolidated group-level and on individual banks/MCIs; insurance tests cover five life insurers, six occupational pension funds, five non-life insurers (each accounting for more than 70 percent in their sector); ATP included.
- Illustrative five-year scenarios (not forecasts; associated with uncertainty):
  - “COVID central”:
    - GDP: sharp fall of 8 percent in 2020, rebound of 6 percent in 2021.
    - Unemployment rate: rises to 6.8 percent in 2020, then gradually declines.
    - Assumption: focuses on growth and unemployment implications; does not entail abrupt tightening of financial conditions or a real estate downturn.
  - “COVID prolonged”:
    - GDP: decline of 8 percent in 2020, further contraction of 2.1 percent in 2021, recovery in 2022.
    - Unemployment rate: rises to 11.8 percent in 2022 before steadily falling.
    - Assumption: no sharp tightening in financial conditions or decline in house prices.
  - “Market shocks”:
    - Characterized by a sharp rise in asset risk premia, adverse effects on banks’ funding costs, and a prolonged domestic real estate downturn amid a protracted recession that is less severe in the short term relative to the other two scenarios.
    - Note: designed before COVID-19; incorporates a combination of credit, market, and cost-of-funding shocks.
- Methodology: same methodology used across scenarios to quantify implications on bank solvency. Tests use 2020 Q1 capital ratios (latest available data at time of testing).

### Solvency stress tests specifics
- Focus: SIFIs consolidated at group-level — seven SIFIs (three financial groups, two banks, two MCIs).
- For financial groups: tests performed on individual banks and MCIs and consolidated by accounting for intra-group holdings.
- Banks: scenarios expected to increase loan loss rates with significant variations across economic sectors and banks; cost-of-funding shocks, only partially passed to borrowers, would further erode net interest income.
- MCIs: assessed capacity to withstand losses through three channels:
  - (i) credit losses triggered by mortgage defaults,
  - (ii) losses from downward repricing of securities held in capital centers,
  - (iii) loss of net interest income from issuance of higher-cost junior bonds needed to shore up collateral of capital centers.
- Assumption: within-group guarantees and guarantees provided by banks outside the stress testing perimeter are available to cushion MCI capitalization impacts.

*Source: 1dnkea2020001 - 5.      Denmark’s financial system is large, with assets over 630 percent of GDP.*

### 20.      The illustrative scenarios reveal a large impact on capital which varies across SIFIs:

### 20.      The illustrative scenarios reveal a large impact on capital which varies across SIFIs:

### Stress test outcomes — capital scenarios
- COVID central scenario:
  - Aggregate CET1 ratio declines from 18.3 percent in 2019 to a low point of 13.4 percent in 2020.
  - Aggregate leverage ratio declines to 3.6 percent.
  - Main driver of losses: materialization of credit risk, predominantly influenced by GDP growth and the unemployment rate.
  - All SIFIs meet their minimum capital requirements (a 4.5 percent CET1 ratio).
  - One out of the seven SIFIs must partially tap into its capital conservation buffer (CCB) to fully absorb the losses under stress.
  - The leverage ratio for one SIFI falls below the three percent threshold (the leverage ratio is currently planned for implementation in mid-2021).

- COVID prolonged scenario:
  - Aggregate CET1 ratio falls to a low point of 12.4 percent in 2022.
  - Aggregate leverage ratio would decline to 3.0 in 2024.
  - All SIFIs still meet their minimum capital requirements, but the impact on capital is more differentiated and subject to greater uncertainty.
  - One SIFI makes partial use of its CCB; another SIFI needs to partially use its SIFI buffer.
  - The leverage ratios of a few SIFIs fall below three percent.

- Market shocks scenario:
  - Aggregate CET1 ratio decreases to 13.4 percent in 2024 (trough at the end of the stress testing horizon).
  - Aggregate leverage decreases to 3.4 percent.
  - Scenario assumes significantly more severe market shocks and a lasting domestic real estate downturn; results are primarily driven by the protracted nature of the envisaged downturn.
  - A few SIFIs make partial use of their CCBs and a few SIFIs’ leverage ratios dip slightly below three percent.

### Interpretation and limitations of stress tests
- Non-linearity and second-round effects:
  - Relationship between growth and capitalization is non-linear, implying deeper recessions can have disproportionate impacts on capital.
  - Some second-round and non-linear effects are not fully captured and could imply an underestimation of losses.
- Assumptions and caveats:
  - Analysis presumes pre-crisis relationships between macroeconomic and financial variables remain unchanged.
  - Possible amplification of COVID shocks for banking liquidity or potential system-wide contagion risks have not been assessed.
  - Further deterioration of economic prospects and tighter financial conditions, relative to those assumed under the scenarios, could bring about a situation where SIFIs breach their minimum capital requirements.

### Liquidity stress tests — key findings
- Tests assessed resilience to severe shocks to cash inflows/outflows combined with asset liquidations subject to valuation haircuts.
- Assumed shocks were significantly larger than those that materialized in 2020 Q1.
- Banks can counterbalance negative funding gaps by using cash and by liquidating securities both in the markets and at the central bank (via standard monetary operations).
- Results:
  - All five banks would be able to survive extreme liquidity pressures without external support for longer than three months.
- Risks:
  - Tighter financial conditions relative to those assumed could affect banks’ access to liquidity and funding costs unevenly and may result in acute liquidity strains for some banks.
- Note: The liquidity stress tests were conducted before the onset of the COVID-19 pandemic.

### Insurance sector stress tests — design and outcomes
- Methodology:
  - Resilience assessed through top-down (TD) and bottom-up (BU) solvency stress tests implemented before the onset of the COVID-19 pandemic.
  - Tests build on the narrative and severity of the market shocks adverse scenario and encompassed several market shocks combined into a stress scenario considerably more severe than observed in 2020 Q1.
  - Market shocks are the main risk factor for the insurance sector, especially life insurance; credit risk is more critical for banks.
- Aggregate solvency outcomes under the adverse scenario:
  - Regulatory solvency requirements are met in aggregate, but the life insurance sector is significantly impacted.
  - BU tests:
    - Median solvency capital requirement (SCR) ratio in the life sector dropped from 188 to 127 percent after stress.
    - Median SCR ratio in the non-life sector declined from 233 to 207 percent after stress.
    - One life company drops below a solvency ratio of 100 percent.
    - Capital needed to restore full coverage of solvency requirements would be only a moderate fraction of the sample's aggregate capital.
  - TD tests broadly confirmed BU results.
- Risks:
  - Resurgent financial market turbulence associated with even more severe shocks relative to those assumed could impair insurers’ solvency positions and profitability to a greater extent.

### Medium-term insurance sector implications and responses
- Investment returns and profitability:
  - Over the medium-term, insurers will face declining investment returns as their higher-coupon bond holdings expire.
  - In aggregate, the Danish life insurance sector still records positive spreads of investment returns over guaranteed interest rates and is expected to regain profitability after the adverse scenario realization, but significant differences exist across companies.
  - Insurers more active in non-life and unit-linked life business are less affected by the low-yield environment.
  - Companies with a high legacy stock of guarantees are likely to experience a drain on profitability.
- Portfolio rebalancing under stress:
  - Life insurers would consider gradually divesting riskier holdings (equities, non-investment grade bonds) and increase investments in sovereign and covered bonds.
  - Non-life companies would rebalance portfolios and may expand investments in equities and real estate.
  - BU survey responses suggest insurers’ reactions under the adverse scenario may mitigate the impact of shocks on financial markets as they try to shore up capital buffers.

### Sensitivity tests — longevity and climate-related risks
- Longevity risk:
  - A strong increase in life expectancy would reduce the median SCR of life insurers and occupational pension funds to 210 percent.
  - Higher mortality rates, including temporarily from a severe epidemic, would decrease annuity payments; coverage implications for sickness pay and hospitalization are only moderate.
- Climate-related catastrophic events:
  - Non-life insurers mainly exposed to windstorms and cloudbursts; cloudbursts occur more frequently but are local with moderate losses per event.
  - A repetition of windstorm Anatol (1999) would result in a reduction of non-life insurers’ own funds of only one percent, as risks are largely ceded to reinsurers.
- Note: These sensitivity tests were conducted prior to the COVID-19 pandemic.

### ATP stress test outcomes
- ATP displays a considerable degree of resilience in the stress test.
- Its individual reserve requirement remains covered by a significant margin.
- Stress impacts:
  - Stress reduces ATP’s profits, which would turn negative in the first year of the projection horizon and remain flat in the following two years assuming asset prices do not recover.
  - Wider spreads and a sharper decline in asset prices relative to those assumed under the market shocks adverse scenario could reduce ATP’s bonus potential (free reserves) to a larger degree.
- Note: The stress test of ATP was implemented before the COVID-19 outbreak.

### Interconnectedness and contagion analysis — banking system structure and transmission
- Methodology:
  - Novel multi-layer contagion model distinguishing transmission of shocks across eight different exposure types based on the new credit registry database.
- Key findings:
  - MCIs play a central role in the domestic interbank system and take center stage in the network linking systemic and non-systemic banks through a dense network of covered bond exposures.
  - The full network highlights strong connections between the Danish banking system and:
    - Domestic corporate sector (in particular, CRE) and household sectors via loans.
    - Domestic institutional investors, including ICPFs and investment funds, via securities.
    - Cross-border exposures are especially strong with the Nordic region and the euro area financial system.
  - Danish credit institutions are mostly exposed to shocks from within the banking system:
    - MCIs are a key source of outward spillovers and induce the highest levels of contagion losses through unlisted shares (reflecting complex group structures) and covered bond exposures.
    - Contagion is transmitted predominantly through credit losses.
    - Systemic banks, on average, are more vulnerable to inward spillovers owing to their covered bond holdings.
    - Analysis identifies four financial institutions with the ability to amplify spillovers across the financial system given their high contagion and vulnerability scores.
    - Losses from unlisted shares occur mostly within group structures, while potential contagion from covered bond exposures can be more widespread.

### Macroprudential policy — institutional setup, weaknesses, and recommendations
- Institutional features:
  - Macroprudential authority: Systemic Risk Council (SRC), chaired by the chairman of the Board of Governors of the DN.
  - SRC role: advisory only, can issue observations, warnings, and recommendations with an explicit “comply-or-explain” mechanism.
  - Ultimate decision-making power rests with the minister of industry, business and financial affairs (MIBFA); in practice, minister needs support of the economic committee, so hard powers effectively lie with the government.
  - DN plays an important role in systemic risk monitoring; SRC secretariat is housed in the DN and prepares SRC quarterly discussion materials.
  - SRC secretariat has limited publications due to resource constraints; more frequent publications would enhance communication and awareness.
- Actions taken since 2014 FSAP to address household vulnerabilities:
  - Mandatory down-payment requirement and other demand-side measures targeting loan origination and amortization.
  - Limits on lending via interest-only and floating-rate mortgages to highly indebted households.
  - These measures, alongside property tax reforms and easing supply constraints, appear to have helped arrest rapid property price increases.
- Countercyclical capital buffer (CCyB):
  - CCyB was increased to two percent (effective December 2020) and fully released following the onset of the COVID-19 pandemic.
  - Important that credit institutions use provided flexibility to extend lending capacity and not pay additional dividends or conduct share buybacks.
- Sources of inaction bias and framework limitations:
  - SRC legal framework emphasizes striving for consensus; secretariat starts drafting proposals only after mandate from SRC, often following consensus building — this can delay decision-making and confer implicit veto power to members.
  - SRC has limited hard powers; MIBFA did not comply with SRC’s recommendation on deferred amortization loans in 2014 and only partially complied in 2017, contributing to inaction bias and adoption of second-best policy options.
  - Lack of legal basis to introduce binding borrower-based macroprudential measures—many demand-side measures are implemented as guidelines under consumer protection legislation rather than binding macroprudential tools.
- Policy recommendations to strengthen macroprudential framework:
  - Give the SRC chair, enshrined in law, the ability to make proposals for a recommendation—after due consultation with other SRC members—without the need to strive for consensus, to limit delays before the secretariat starts working on draft proposals.
  - Over time, if inaction bias persists, consider giving SRC hard powers over capital tools under CRD/CRR legislation (make SRC the designated authority for macroprudential policy). Borrower-based tools can remain with MIBFA with the SRC having recommendation powers and a comply-or-explain mechanism.
  - Introduce national legislation to include borrower-based tools (limits on LTV, DTI, and DSTI) in the policy toolkit to enhance legitimacy of macroprudential policy.
  - Over the medium term, introduce a stricter LTV limit (at least 90 percent) to safeguard against large house price shocks.
  - If the uptrend in house prices continues, consider introducing binding income-based limits, such as DTI restrictions for all loans, irrespective of their LTV ratios.
- Addressing structural housing market distortions:
  - Structural factors to be addressed include:
    - Favorable tax treatment of owner-occupied housing compared to other saving vehicles.
    - High mortgage interest deductibility causing debt bias alongside tax exemption on capital gains in owner-occupied housing.
    - Complex rental market regulations with rent caps across a significant share of apartment buildings in major cities, creating a lack of housing supply.
  - Ministries’ participation in SRC should assist complementary use of fiscal measures for both fiscal and macroprudential policy.
- CRE market monitoring:
  - SRC should closely monitor CRE market risks and stand ready to act if vulnerabilities intensify.
  - Enhance coverage and quality of CRE data, including centralized collection of market data and more granular information on CRE financing.
  - If CRE risks continue to rise, consider macroprudential measures such as increasing risk weights on CRE exposures or introducing a sectoral systemic risk buffer.
- Enhancing SRC monitoring framework:
  - Further expand coverage to nonbank sectors.
  - Develop truly macroprudential stress tests that take into account macro-financial feedback loops.
  - Build on interconnectedness work, particularly given the centrality of covered bonds in the system.

*Italic: IMF staff summary of chapter content.*

### 41.      The Danish Financial Supervisory Authority (DFSA) has improved standards in its

### 1dnkea2020001 - 41.      The Danish Financial Supervisory Authority (DFSA) has improved standards in its

### DFSA: overall assessment and recent changes
- The DFSA’s supervisory approach focuses on traditional financial risks; in particular credit risk in banking and market risks and asset-liability management in insurance.
- The DFSA benefits from and uses a suite of corrective and enforcement powers, which were further augmented through AML/CFT legislation (that entered into force in January 2020).
- Since the last FSAP the DFSA has improved standards in its oversight of banking and insurance.

### Operational independence and governance
- The DFSA’s operational independence is compromised through governance arrangements and through resource constraints.
- The funding structure: fees levied on the industry but budget set by the MIBFA (which has a non-voting representative on the Board) does not isolate the DFSA from potential influence of government or industry.
- Recommendation: The MIBFA representative should be removed from the DFSA Board and the reasons for the dismissal of a Director General of the DFSA should be clearly stated in law and publicly disclosed.
- The requirement that the DFSA operate on a “least cost” basis raises concerns that it undermines quality and further development of policy, processes, scope and execution of tasks.

### Resources and capacity building
- Risks to the industry, including money laundering, are better understood than previously, though supervisory skills and resources to mitigate those risks are still being built up as newly hired AML/CFT and insurance inspectors are still being trained.
- The DFSA would benefit from a further increase in resources.

### Banking supervision: enhancements and gaps
- Enhancements made since last FSAP:
  - Inspection cycle interval for sound and well-capitalized medium and smaller banks shortened from once every six to four years.
  - A major project completed to document the DFSA’s business processes.
- Remaining needs:
  - The business-process documentation project needs to be complemented by a similar project to support consistency of analytical approaches and risk interpretations.
  - Enhanced approaches to confirm explicitly the veracity of supervisory data received, as part of standardized reporting schemes, are needed.
  - Nonfinancial risks warrant greater prominence: entities’ compliance, governance, conduct, compensation practices, and risk culture require sharper supervision.

### Insurance supervision: Solvency II effects and outstanding issues
- Implementation of the Solvency II Directive has enhanced insurance supervision and improved risk culture across the sector.
- The Directive addresses many 2014 FSAP recommendations and has raised the level of observance with the Insurance Core Principles.
- Remaining needs:
  - Increase on-site inspection frequency, particularly in the non-life sector where inspections remain too infrequent.
  - Complete a solid risk assessment framework that includes non-financial risks, in particular governance-related risks.
  - Supervision of cross-border business should be strengthened after two smaller non-life insurers failed in 2018 due to insufficient reserves and incorrect asset assessment.
  - A new supervisory standard is being developed to address previous gaps in licensing and supervisory monitoring.

### Financial integrity (AML/CFT)
- Reforms following the 2019 Article IV Consultation and the Danske Bank case:
  - Recent legislation has significantly strengthened the DFSA’s AML/CFT monitoring and enforcement powers, notably by allowing it to levy administrative fine notices and stiffening existing penalties for non-compliance with certain AML/CFT obligations.
  - Since June 2019, the DFSA’s AML Division has expanded considerably, enabling an increasing number of on-site inspections.
  - The DFSA’s AML Division has developed, but not yet finalized, a new institutional risk assessment model.
  - The MLF authorities have adopted a new MOU to guide coordination of domestic AML/CFT activities.
- Recommendations and priorities:
  - Complete testing and refining of the DFSA’s new AML/CFT risk model, issue the associated final questionnaires, and use results to set the annual on-site inspection schedule.
  - Intensify on-site inspections of higher-risk financial institutions and demonstrate effective use of new monitoring and enforcement powers in response to violations.
  - Strengthen international cooperation arrangements: Government should consider next-stage options for integration/consolidation of AML/CFT supervision at sub-regional or EU levels and pursue selected option(s).
  - Strengthen domestic cooperation arrangements, including public–private sector interaction: the MLF should review recommendations released by Finance Denmark’s AML Task Force that would imply action by the authorities, and advance agreed initiatives.

### Systemic liquidity and covered bonds
- Covered bond markets key to bank funding and mortgage financing; they demonstrated resilience during past turbulence but important tail risks remain.
- Recent market behavior (March–April 2020):
  - After a brief period of subdued primary activity in March 2020, Danish covered bond issuers resumed new issuances and facilitated mortgage financing.
  - In the secondary market, covered bond spreads spiked during March and early April, but recovered quickly thereafter.
- The high degree of interconnectedness through covered bond holdings means ensuring health of this market under stress is crucial for financial stability.
- DN (Danmarks Nationalbank) operational framework:
  - DN's standard operational framework is credible and effective; DN has kept the exchange rate stable through policy rate adjustments and FX interventions.
  - Introduction of daily market operations in 2017 has improved functioning of the interbank market by reducing short-term interest rate volatility.
- DN’s crisis response:
  - DN promptly provided liquidity support during the crisis, launching extraordinary liquidity facilities and reactivating bilateral swap lines with the ECB and the Federal Reserve.
  - These measures were accompanied by stabilization in money markets and easing of liquidity and financial conditions.
- Operational preparedness improvements recommended:
  - Complete planned refinements to framework for accepting non-standard collateral (including credit claims). Increasing automatization for mobilization, eligibility assessment, and pricing of credit claims would significantly improve operational efficiency and broaden collateral base for market-wide liquidity support and ELA.
  - Seek greater domestic interagency information sharing and collaboration to enhance operational preparedness for liquidity support (for example, via joint simulation exercises and horizon scanning).
  - Communicate clearly to key financial institutions the specific information requirements in the context of ELA.

### Crisis management and financial safety nets
- Since 2014, Denmark’s financial safety net and crisis management frameworks, including bank resolution, have improved significantly via EU transposition and domestic reforms:
  - New legislation for resolution and deposit insurance; introduction of resolution framework for banks and MCIs; designation of two national resolution authorities (DFSA and the Financial Stability Company, FSC); governance changes to the deposit insurance system (DIS); revival of cross-border cooperation through the Nordic-Baltic Stability Group (NBSG); joint crisis simulation in 2019.
  - Denmark’s resolution strategy for small- and medium-sized banks has been successfully tested in practice.
- Governance recommendations for resolution authorities, especially the FSC:
  - Strengthen operational autonomy so the FSC Board can execute mandate without interference from government or industry.
  - Involve the government in resolution decisions only when fiscal support is needed.
  - Board members should serve a term disconnected from the political cycle and sufficiently long to develop expertise.
  - Formal dismissal procedure should limit dismissals to certain causes and require a statement of reasons; prohibit active bankers from serving as FSC Board members to prevent perceived conflicts of interest.
- SIFIs (systemically important financial institutions):
  - Further progress needed to ensure operational readiness to rapidly execute recovery and resolution measures.
  - Resolution planning for SIFIs should be expedited, prioritizing operational continuity and conceptual work on use of a bridge MCI and practical execution of the bail-in tool.
- Liquidity for institutions in resolution:
  - The COVID-19 pandemic highlighted the need to define strategies for liquidity assistance for institutions in resolution.
  - DN should be able to provide liquidity to enable resolution plans to be put into effect, subject to safeguards, to an institution considered systemic and viable whose solvency concerns will be resolved with certainty and within a short time frame.
  - Authorities need to assess if the Resolution Fund (with back-up funding from the government) is well suited to secure needed amounts within a short period of time and on a continuous basis.
- Other contingency planning and crisis management enhancements:
  - The Coordination Committee for Financial Stability needs to include the FSC as an important pillar of the financial safety net.
  - Under the Coordination Committee, nationally coordinated crisis management and communication plans should be developed.
  - Authorities should test operational readiness with a system-wide financial crisis simulation exercise.
  - Note: The Coordination Committee consists of high-level representatives from the MIBFA, MOF, DFSA, and DN and its tasks include coordinating the failure of a SIFI or coordination of financial crisis policy responses.

### Boxed summaries from the source (key policy measures, Danske Bank case, mortgage finance)
- Box 1: Key policy measures taken in response to the COVID-19 pandemic (as of June 2020)
  - Fiscal policies:
    - Discretionary fiscal support of DKK 60 billion (2.6 percent of 2019 GDP) initially; increased spending mainly to finance health care needs and extraordinary budgetary measures to support workers and businesses.
    - Another 6.2 percent of GDP in countercyclical support expected to come through automatic stabilizers.
    - Additional DKK 30.7 billion in fiscal support (1.3 percent of 2019 GDP) and about DKK 70 billion (3.0 percent of 2019 GDP) in guarantees and liquidity measures announced later.
    - Green renovation of public housing (DKK 30.2 billion or 1.3 percent of 2019 GDP) planned for 2021-16.
  - Monetary and financial policies:
    - Danmarks Nationalbank (DN) announced on March 12 the launch of an “extraordinary lending facility” later expanded to a three-month collateralized lending operation.
    - Standing swap line with ECB was activated and its size was doubled to EUR 24 billion.
    - Agreement with the Federal Reserve to establish a USD 30 billion swap line that will stand for at least 6 months.
    - DN increased the policy rate by 15 basis points to –0.6 percent; monetary policy rates remain lower in Denmark than in the euro area. DN maintains the krone’s tight peg to the euro.
    - Countercyclical capital buffer was preemptively released and planned increases canceled (March 12).
    - DFSA announced a case by case relaxation of regulation on the liquidity coverage ratio (LCR) requirement; DFSA urged banks and insurance companies not to pay dividends or buy back shares (March 30).
- Box 2: The Danske Bank case (summary)
  - Possible laundering of up to $230 billion in payments through the Estonian branch of Danske Bank from 2007-2015.
  - Danske Bank acquired Finnish Sampo Bank in 2007, converting it into a branch of Danske Bank A/S in 2008.
  - From 2007-2015, the Estonian branch serviced approximately 15,000 non-resident customers in its International Banking Department and other units.
  - External parties transferred approximately DKK 1,500 billion ($230 billion) to the Estonian branch’s non-resident customers, who then forwarded a similar sum to external recipients.
  - JP Morgan terminated its correspondent banking relationship with the Estonian branch in 2013; Bank of America and Deutsche Bank’s U.S. subsidiary followed suit in 2015.
  - In December 2017, Danske Bank announced it had been fined DKK 12.5 million ($2 million) by the Danish authorities for violating AML/CFT rules related to monitoring of transactions to and from correspondent banks.
  - In September 2018, Danske Bank’s internal investigation concluded a large proportion of the transfers were suspicious and potentially illegal, and that AML/CFT procedures at its Estonian branch had been “manifestly insufficient and inadequate”.
  - In January 2019, the DFSA released a report on its supervision of Danske Bank as regards this case.
  - In February 2019, the EBA’s Board of Supervisors set up a panel to investigate possible breach of EU law in connection with DFSA’s and EFSA’s supervision of Danske Bank and its Estonian branch.
  - In April 2019, the EBA BoS acknowledged past failings of DFSA and EFSA’s supervision but rejected the panel’s recommendation with respect to the breach of union law proceeding.
  - In October 2019, Danske Bank announced its Estonian business had entered into liquidation in response to an EFSA order.
- Box 3: Key elements of the Danish mortgage finance system
  - MCIs’ business activities are limited to provision of mortgage loans (legally not allowed to accept deposits).
  - Loans must be funded by issuance of covered bonds backed by assets segregated in bankruptcy remote capital centers (still on the MCI balance sheet), and a residual claim (ranking senior to non-covered bond creditors) on the general balance sheet if capital center assets cannot fully cover investor exposure.
  - A margin over the covered bond cost is charged to cover administrative costs, risk, and profit.
  - All issuers required to apply a balance principle which serves as a strict asset-liability management mechanism: mortgage loan-related cash flows are passed through to the covered bond investor, reducing issuers’ exposure to market risks.
  - Refinancing risk attenuated by legislation allowing extension of a bond’s maturity by one year if a refinancing auction fails or if interest rates have risen by more than five percentage points in the preceding year.

*Source: IMF staff summary of the chapter content provided.*

### Box 4. Denmark: Climate Change and Financial Stability in Denmark

### Box 4. Denmark: Climate Change and Financial Stability in Denmark

### Physical risk: flood exposure
- Flood-related disasters are identified as one of the main physical risks associated with climate change confronting Denmark.
- Many Danish cities and large towns are located along the coastline with many buildings situated in low-lying areas, rendering a sizeable share of banking assets prone to flood-induced risks.
- Danmarks Nationalbank (2019) estimates:
  - Currently "1½ percent" of mortgage assets serving as collateral are at risk of flooding, but this share could increase to "13 percent".
- Acknowledging uncertainty in estimates, current assessments indicate:
  - "1-3 percent" of SIFIs’ mortgaged assets are at risk of flooding;
  - For two SIFIs, the estimate could exceed "15 percent" reflecting greater exposure to rising sea levels.

### Transition risk: carbon intensity and sectoral exposures
- The financial sector has relatively limited exposure to the highest carbon-emitting industries.
- Credit institutions have generally provided lending to corporate customers in industries with less carbon-intensive production processes, accounting for "23 percent" of corporate sector lending.
- Transition risks are larger for institutions with large exposures to:
  - the agricultural sector (especially relevant for smaller banks), and
  - the energy sector.

*Source: Danmarks Nationalbank (2019), “Climate Change Can Have a Spillover Effect on Financial Stability, Analysis No. 26 (December). *

### 1. Prolonged

### 1. Prolonged COVID-19 outbreak

### Risk overview and macrofinancial transmission
- Containment measures remain in place (in some places intensify or need to be re-introduced) through early 2021.
- Longer containment and uncertainties about the intensity and the duration of the outbreak reduce supply and domestic and external demand, with sustained negative impact on Danish exports, investment, and consumption.
- Pandemic-prompted protectionist actions could exacerbate adverse effects given Denmark is a small open economy with many sectors—including shipping—deeply integrated in global value chains.
- Corporate-sector losses would be accompanied by further increases in unemployment, NPLs, and loan loss impairments weighing on bank profitability, possibly aggravated by tighter credit conditions and adverse Nordic spillovers.
- Deteriorating fundamentals and declining risk appetite could trigger financial tightening and a macrofinancial feedback loop: weaker domestic demand → declines in asset prices → erosion of household wealth → weaker consumption → credit crunch, amplified by elevated household indebtedness and Nordic financial integration.
- International investors could retrench from the Danish covered bond market, raising funding costs and creating (re-)financing problems, especially for MCIs.

### Selected risks from the Risk Assessment Matrix
- Financial shocks originating in the Nordic-Baltic region (Overall Level of Concern: Medium)
  - Protracted market volatility, tighter regional financial conditions, or a correction in real estate markets in other Nordic countries would trigger losses for major Danish SIFIs due to large MCI subsidiaries.
  - Interlinkages with Nordic and Baltic countries amplify effects of global trade and financial shocks; second-round effects arise from tight regional trade ties, financial interlinkages, and cross-holdings of covered bonds.
  - Ongoing (regional) ML cases and significantly larger-than-anticipated ML-related fines by foreign authorities could disrupt banking groups’ capacity to conduct business and hurt profitability.
- Prolonged downturn in the domestic real estate market (Overall Level of Concern: Medium)
  - A decline in real estate prices would drag down other asset classes, dampen consumption and investment (especially leveraged households and non-diversified real estate developers), increase unemployment, and raise NPLs and impairment losses, potentially creating a vicious cycle of tighter financial conditions and depressed activity.

### Banking sector stress testing: institutional perimeter and market shares
- Institutions included (solvency stress tests):
  - 7 systemic credit institutions: Danske Bank Group, Jyske Bank Group, Nykredit Realkredit Group, Sydbank, SparNord Bank, DLR Kredit, Nordea Kredit.
  - 5 commercial banks (unconsolidated): Danske Bank, Nykredit Bank, Jyske Bank, Sydbank, Spar Nord Bank.
  - 6 mortgage credit institutions (unconsolidated): Realkredit Danmark, Nykredit Realkredit, Totalkredit, Jyske Realkredit, DLR Kredit, Nordea Kredit.
- Market share:
  - The 7 largest systemic credit institutions cover approximately 77 percent of the total assets for credit institutions.
  - The 5 largest banks cover approximately 84 percent of the banking system’s total assets.
  - Active nodes in contagion analysis account for 86 percent of the total assets of the Danish banking system.
- Data and baseline dates:
  - Baseline Date (solvency tests): 30 September 2019.
  - Latest data (liquidity tests): August and September 2019.
- Stress test horizon: Five years (2020–2024).

### Channels of risk propagation and modeling approaches
- Approach: Balance sheet-based top-down tests, with macro-financial satellite models for credit, market, and net interest income channels.
- Credit risk (banks): Loss calculations based on a panel regression model for logit transformed sectoral impairment rates; regressors include lagged dependent variable, unemployment rate and its first lag, GDP growth rate and its first lag, short term interest rate and its first lag.
- Market risk: Market losses from holdings of debt instruments calculated via full re-valuation of each individual security incorporating hedges; IMF provided yield-curve shocks and DN stress team performed calculations.
- Net interest income: Gap analysis based on granular asset/liability structure broken into funding sources and re-pricing buckets; cost of funding shocks defined by liability type.
- MCIs:
  - Affected by: 1) materialization of credit risk; 2) materialization of market risk leading to downward repricing of securities held in capital centers; 3) net interest income impairment when MCIs issue junior bonds to top-up collateral at a negative spread vis-a-vis sovereign bonds.
  - Credit risk for MCIs: Time series regression for logit transformed aggregate impairment rates; regressors include lagged dependent variable, GDP growth and its lag, unemployment and its lag, short-term interest rate and its lag, inflation and its lag.
  - Top-up collateral needs: Triggered when housing prices (index) decline and LTV ratios rise above the (80/60 percent limits, depending on mortgage type), forcing MCIs to issue bonds to restore LTV thresholds.
- Credit growth assumption: Assumed to grow with nominal GDP growth rate, floored at 0.

### Scenario analysis and key numeric scenario outcomes
- COVID central scenario:
  - Sharp fall in GDP of 8 percent in 2020, followed by a rebound of 6 percent in 2021.
  - Contraction in GDP for 2020 is worse than projected in the IMF WEO for April.
  - Cumulative decline of real GDP relative to the September 2019 WEO projections equivalent to 1.8 standard deviations (6.1 percentage points) over two years.
  - Unemployment goes up to 6.8 percent in 2020, followed by gradual recovery reaching 5.6 percent in 2023.
  - Cost of funding shocks for 2020: 30 basis points for interbank lending and riskier debt securities and 15 basis points for covered bonds and repos.
- COVID prolonged scenario:
  - Instead of a rebound in 2021, the economy continues to contract by 2.1 percent before recovery ensues in 2022.
  - Cumulative decline of real GDP relative to the September 2019 WEO projections equivalent to 4 standard deviations (13.6 percentage points) over two years.
  - Unemployment rate peaks at 11.8 percent in 2022.
  - Cost of funding shocks for 2020: same calibration as COVID central (30 basis points and 15 basis points).
- Market shocks adverse scenario:
  - Triggered by combination of external and domestic shocks generated using IMF’s Flexible System of Global Models.
  - Escalating global trade tensions and tightening global financial conditions depress confidence across advanced and EM economies.
  - Abrupt decompression of risk asset premia raises money market interest rates, banks’ short-term funding costs, and long-term government bond yields.
  - Housing prices decline by 18 percent over three years, depressing domestic consumption and residential investment.
  - Overall cost of funding increases by 100 basis points, with partial pass-through to lending rates.
  - Adverse scenario results in cumulative decline of real GDP relative to the October 2019 WEO projections equivalent to 2.5 standard deviations (8.6 percentage points) over two years.
- Yield curve shocks for market risk loss calculations are broadly in line with observed COVID-induced stress.

### Behavioral adjustments and reporting conventions in stress tests
- Balance sheets assumed quasi-static: loan growth and funding growth tied to nominal GDP growth.
- No asset disposal/acquisition; external capital injections not permitted.
- Non-performing loans do not generate interest income.
- Dividend policy assumption: Banks assumed to make dividend pay-outs of 40 percent for periods with positive net income and no pay-outs in case of negative income.
- Tax calibration: Banks pay the same tax rate calibrated as the average tax rate for the five large banks during the last four quarters; same calibration for MCIs.
- Sensitivity analysis:
  - Static shocks to single risk factors.
  - Simultaneous default of 3, 5, and 10 largest single exposures with several LGD assumptions.
  - Significant increase to RWAs as a result of Basel-III output floors.
- Reporting output: Evolution of CET1 capitalization ratios for consolidated banking groups; decomposition of CET1 capital ratio reduction by drivers (credit risk, market risk, interest rate risk).

### Banking sector liquidity stress testing
- Institutions included (liquidity tests): 5 commercial banks (unconsolidated): Danske Bank, Nykredit Bank, Jyske Bank, Sydbank, Spar Nord Bank.
- Methodology: Top-down cash-flow-based tests using time structure of undiscounted cash flows for up to one year, by currency; Top-Down LCR based tests.
- Tail shock size:
  - Bank run on deposits and partial dry up of wholesale funding markets, including the covered bonds market.
  - Haircuts applied to liquid assets to capture fire-sale effects.
  - Run-off rates calibrated to trigger a severe cumulative withdrawal of overall funding amount (about 20 percent over a three-month period).
- Buffers: Counterbalancing capacity; DN (standing) facilities (general framework only, excluding unconventional facilities).
- Reporting output: Liquidity gap by bank; survival period by bank and number of banks that still can meet their obligations.
- Regulatory standards: Liquidity gap and survival period consistent with Basel III standards (LCR).

### Interconnectedness and contagion analysis
- Institutional perimeter:
  - Active nodes: Systemic banks (Danske, Jyske, Nykredit, Sydbank, Spar Nord), non-systemic banks (Arbejdernes, Ringkjobing, Sk. Kronjylland, Sk. Sjaelland-Fyn, Lan & Spar, Vestjysk, Sk. Vendsyssel, Jutlander, Den Jyske), mortgage credit institutions (Nykredit Realkredit, Realkredit Danmark, Totalkredit, Nordea Kredit, Jyske Realkredit, DLR Kredit, LR Realkredit).
  - Passive nodes: 109 “Group 3” banks and branches of foreign banks, 103 life and non-life insurers, 41 pension funds, aggregate sectors (investment funds, other financial institutions, households, NFCs by industry), and cross-border counterparties aggregated by country and sector.
- Data and baseline date: Credit registry, securities database, supervisory COREP/FINREP/national reporting templates; single datapoint September 30, 2019.
- Methodology: CoMap framework extended to model multi-layer contagion through 8 instrument types (loans, deposits, reverse repos, other claims, covered bonds, other bonds, equities and unlisted shares) and to incorporate MCIs as active nodes.
- Risk channels assessed:
  - Credit default channel: Simulates entity default with losses calculated based on exposure-level LGD; for equities and unlisted shares LGD assumed 100 percent.
  - Funding withdrawal channel: Simulates funding shortfalls met by HQLA in excess of NLO or leading to deleveraging and fire sales; uses exposure-specific maturity and asset encumbrance data.
  - Market repricing channel: Captures market reactions to weakening capitalization of MCIs and banks via rating downgrades and valuation losses to holders of downgraded securities.

*Source: 1dnkea2020001 - 1. Prolonged (PDF chapter).*

### 3. Tail ShocksScenario analysis •Idiosyncratic shocks: hypothetical failure of each entity as a potential trigger event.

### 1dnkea2020001 - 3. Tail ShocksScenario analysis •Idiosyncratic shocks: hypothetical failure of each entity as a potential trigger event.

### Tail shocks — scenario analysis (general)
- Adverse scenario shock assumptions:
  - Risk-free interest rates (without volatility adjustment): +/-0 bps (1y DKK), +25 bps (10y DKK)
  - Sovereign bond spread: +110 bps (Denmark)
  - Sovereign bond spread: +40 bps for AAA-rated euro area countries
  - Sovereign bond spread: up to +250 bps for high-yield euro area countries
  - Stock prices: -26.5 percent (domestic)
  - Stock prices: -20 percent (other advanced economies)
  - Stock prices: -25 percent (emerging and developing economies)
  - Property prices: -22 percent (domestic residential)
  - Property prices: -26.4 percent (domestic commercial)
  - Property prices: -7.7 and -9.2 percent (other countries, residential and commercial, respectively)
  - Domestic corporate bond spreads (financials): between +100 bps (AAA) and +300 bps (B and lower)
  - Domestic covered bond spreads: +80 bps (AAA)
  - Foreign corporate bond spreads (financials): between +50 bps (AAA) and +300 bps (B and lower)

### Sensitivity analysis (general)
- Sensitivity directions listed:
  - Default threshold
  - Funding shortfall maturity buckets
  - Rating conversion and migration
  - Default ratio for households and NFCs

### Reporting form for results (general outputs)
- Output presentation items:
  - Number of cascade defaults.
  - Contagion losses induced (contagion index) and experienced (vulnerability index) by each node.
  - Decomposition of losses by layer (instrument type) and entity types.

### Appendix II — Insurance Sector Stress Testing Matrix (STeM): domain assumptions (INSURANCE SECTOR: SOLVENCY RISK)

#### 1. Institutional perimeter
- Institutions included (Bottom-Up and Top-Down lists are identical in the source):
  - 5 life insurers: AP Pension, Danica, PFA Pension, Top Liv, Velliv
  - 6 occupational pension funds: PenSam, PK Farmakonomer, PK Socialrådgivere, PK Sundhedsfaglige, PK Sygeplejersker, Sampension
  - 5 non-life insurers: AlmBrand, Codan, Lærerstandens Brandforsikring, TopDanmark, Tryg
  - 1 pension fund: ATP
- Market share:
  - Life and occupational pension funds: >70 percent of total balance sheet assets
  - Non-life: >70 percent of total balance sheet assets
- Consolidation: Solo level
- Data: Regulatory reporting
- Reference date: June 30, 2019

#### 2. Channels of risk propagation
- Methodology:
  - Investment assets: market value changes after price shocks, affecting the solvency position
  - Sensitivity analysis: effect on available capital and solvency position
  - Time horizon:
    - Instantaneous shock
    - 3-year projection (in the scenario)
    - Instantaneous shock

#### 3. Tail shocks — Insurance Sector (detailed adverse scenario)
- Adverse scenario shock assumptions (repeated in insurance STeM):
  - Risk-free interest rates (without volatility adjustment): +/-0 bps (1y DKK), +25 bps (10y DKK)
  - Sovereign bond spread: +110 bps (Denmark)
  - Sovereign bond spread: +40 bps for AAA-rated euro area countries
  - Sovereign bond spread: up to +250 bps for high-yield euro area countries
  - Stock prices: -26.5 percent (domestic)
  - Stock prices: -20 percent (other advanced economies)
  - Stock prices: -25 percent (emerging and developing economies)
  - Property prices: -22 percent (domestic residential)
  - Property prices: -26.4 percent (domestic commercial)
  - Property prices: -7.7 and -9.2 percent (other countries, residential and commercial, respectively)
  - Domestic corporate bond spreads (financials): between +100 bps (AAA) and +300 bps (B and lower)
  - Domestic covered bond spreads: +80 bps (AAA)
  - Foreign corporate bond spreads (financials): between +50 bps (AAA) and +300 bps (B and lower)

#### Sensitivity analysis — Insurance Sector (explicit shocks)
- Longevity shock (life and pensions only): permanent 25 percent decline in mortality rates
- Mortality shock (life and pensions only): permanent 15 increase in mortality rates
- Pandemic event (life and pensions only): temporary 35 percent increase in disability/morbidity rates, temporary 10 percent increase in mortality rates
- Mass lapse event (life only): 20 percent lapse rate in contracts where the discontinuance of the contract results in a loss for the insurer
- Catastrophic events (non-life only): repetition of Windstorm Anatol (December 1999) based on today’s exposures
- Default of largest banking counterparty
- Lower interest rate term structure: -100 bps parallel shift (for all currencies)
- Spread shock to domestic covered bonds: +250 bps

#### 4. Risks and buffers
- Risks/factors assessed:
  - Market risks: interest rates, share prices, property prices, credit spreads
  - Underwriting risks: longevity, mortality, pandemic event, catastrophic events
  - Summation of risks, no diversification effects
  - Credit risks (Top-Down): default of largest financial counterparty
- Buffers:
  - Product-specific
  - Loss-absorption capacity stemming from policyholder participation
- Behavioral adjustments:
  - Management actions limited to non-discretionary rules in place at the reference date (Bottom-Up)
  - None (Top-Down)

#### 5. Regulatory standards and parameters
- Regulatory/accounting standards:
  - Solvency II
  - National GAAP

#### 6. Reporting format for results — Insurance Sector outputs
- Output presentation (Bottom-Up):
  - Impact on solvency ratios (including and excluding the effect of long-term guarantee measures)
  - Impact on net income
  - Contribution of individual shocks
  - Dispersion measures of solvency ratios and net income
  - Survey-based reactions and management actions of insurers after stress
- Output presentation (Top-Down):
  - Impact on solvency ratios (including and excluding the effect of long-term guarantee measures)
  - Contribution of individual shocks
  - Dispersion measures of solvency ratios and net income

*Italic: Source — 1dnkea2020001 - 3. Tail ShocksScenario analysis •Idiosyncratic shocks: hypothetical failure of each entity as a potential trigger event.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2020/english/1dnkea2020001.pdf_