## IMF FSAP — Netherlands (cr1795) ## Source details **Canonical URL:** [IMF FSAP — Netherlands (cr1795)](https://www.imf.org/-/media/files/publications/cr/2017/cr1795.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2017/cr1795.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2017/cr1795.pdf.json) --- ### Macroeconomic context and key vulnerabilities - Gradual domestic recovery amid an uncertain international context. - Key domestic vulnerabilities: - High household debt and prevalent negative equity among young borrowers. - A substantial portion of households have an LTV above 100 percent. - Banks: improved capitalization since the crisis, but contracted balance sheets, low profitability, and significant reliance on wholesale funding. - Ongoing low interest rate environment and slow credit growth weigh on financial institutions. - External and domestic downside risks highlighted: - Euro area recession and deflation, sovereign shocks, tighter global financial conditions, reversal in house prices. - Adverse IMF scenario produces a V-shaped recession with negative growth of 3.7 percent in 2017. ### FSAP focus and adverse scenario design - Emphasis on three “triple threats”: - Low profitability from continued/further decline in interest rates and low credit growth. - High indebtedness of the household and nonfinancial corporate (NFC) sectors. - Banks’ reliance on wholesale funding and exposure to cross-border linkages. - Adverse (extreme but plausible) scenario characteristics: - Cumulative decline of GDP equal to two standard deviations relative to the baseline: 6.9 percentage points over two years and 8.7 percentage points over three years. - Two-year standard deviation of real GDP growth (1980–2015) = 3.4 percentage points. - Layered shocks include: EA tightening, country-specific 15 percent housing price decline, cost of short-term funding rise by 75 basis points, combined funding-cost increase by 100 basis points. - Selected numeric paths (baseline | adverse IMF): - Real GDP growth (percent): Baseline 2015 = 2.0; 2016 = 1.7; 2017 = 1.6; 2018 = 1.9. Adverse IMF 2015 = 2.0; 2016 = 0.1; 2017 = −3.7; 2018 = 0.2. - Short term interest rate (percent): Baseline 2015 = 0.0; 2016 = − 0.3; 2017 = − 0.4; 2018 = − 0.2. Adverse IMF 2015 = 0.0; 2016 = 0.1; 2017 = 0.7; 2018 = 0.7. - Private consumption (percent change): Baseline 2015 = 1.8; 2016 = 1.9; 2017 = 1.9; 2018 = 1.7. Adverse IMF 2015 = 1.8; 2016 = 0.1; 2017 = −4.4; 2018 = −1.0. - Mortgage interest rate (percent): Baseline 2015 = 2.9; 2016 = 2.8; 2017 = 2.7; 2018 = 2.5. Adverse IMF 2015 = 2.9; 2016 = 3.0; 2017 = 3.2; 2018 = 3.0. - Consumer price inflation rate (percent): Baseline 2015 = 0.2; 2016 = 0.2; 2017 = 0.9; 2018 = 1.0. Adverse IMF 2015 = 0.2; 2016 = 0.0; 2017 = −1.4; 2018 = −2.5. - Unemployment rate (percent): Baseline 2015 = 6.9; 2016 = 6.6; 2017 = 6.4; 2018 = 6.1. Adverse IMF 2015 = 6.9; 2016 = 7.2; 2017 = 8.8; 2018 = 8.6. - Fiscal balance ratio (percent): Baseline 2015 = −1.8; 2016 = −1.6; 2017 = −1.0; 2018 = −0.7. Adverse IMF 2015 = −1.8; 2016 = −1.7; 2017 = −1.4; 2018 = −1.5. ### Banking sector — profitability and solvency outcomes - Net interest income (NII) and profitability (baseline projections): - NII for the largest six banks: €29.4 billion (1.3 percent of assets) in 2015; €25 billion in 2017 (about 1.1 percent of assets); €24 billion in 2018 (about 1.1 percent of assets). - Overall banking-system profitability: 0.50 percent of assets in 2015; about 0.15 percent of assets in 2018. - Adverse scenario (CET1 and leverage outcomes): - Fully loaded CET1 for the largest six banks: 13.5 percent in 2015 → 10.2 percent in 2018 (adverse scenario). - Minimum CET1 hurdle rate used: 7 percent over 2016–18; all banks remain above this hurdle—no system capital shortfall on CET1 metric. - Fully loaded leverage ratio for the system (6 largest banks): 3.8 percent → 3.2 percent in the adverse scenario. - One significant bank could fall just below the 3 percent minimum leverage hurdle considered. - Tier 1 capital shortage in the adverse scenario equivalent to €3.8 billion or 0.6 percent of GDP. - Adverse scenario profitability and loss decomposition: - Overall profitability: 0.5 percent of assets in 2015 → −0.3 percent in 2017–18 (adverse). - Net interest income: 1.3 percent of assets in 2015 → 0.8 percent of assets in 2018 (adverse). - Credit loss impairments: rise from 0.2 percent of assets in 2015 to 0.5 percent of assets in 2015 (text phrasing) and decline to 0.3 percent in 2018 as the economy recovers. - Securities valuation: net gain of 0.2 percent of assets in 2015 becomes a 0.2 percent loss in 2017 (due to sovereign yield increases). - PIT PDs peak at 2.4 times the initial level (system-wide) with average multipliers: corporate 2.8; consumer 2.5; mortgages 1.7. - Market-risk losses on securities: duration-adjusted haircuts cause losses equivalent to 2 percent of initial valuations (mitigated by floating-rate securities and allocations to low-spread countries). ### Sensitivity and concentration analyses — key numeric results - Concentration risk (defaults of largest non-sovereign exposures): - No collateral haircut: large banks can withstand simultaneous defaults of up to four largest exposures without undercapitalization (relative to an 8.5 percent Tier 1 hurdle). - 50 percent collateral haircut (extreme): one of the largest banks shows a Tier 1 shortfall upon simultaneous default of the three largest exposures; other SIBs undercapitalized upon simultaneous default of five largest exposures. - RWA-floor sensitivity for mortgage portfolios: - Current IRB mortgage risk weights average 15.4 percent for the largest banks (end-2015). - Floors tested toward standardized 35 percent produce IRB mortgage RWAs of: - 21 percent (60 percent of 35 percent) → CET1 decline equivalent to 0.7 percentage points. - 28 percent (80 percent of 35 percent) → CET1 decline equivalent to 1.4 percentage points. - 31.5 percent (90 percent of 35 percent) → CET1 decline equivalent to 1.8 percentage points. - Tests are static and do not consider phase-in or behavioral responses. ### Liquidity stress tests and maturity-ladder analysis - Cash-flow–based liquidity stress testing calibration and outcomes: - Resulting withdrawals equivalent to 15–20 percent of the initial stock within the first two to three months (variation across banks). - Except for a small institution, all banks could confront persistent and sizable withdrawals without Emergency Liquidity Assistance (ELA) from the ECB for periods longer than six months. - Banks exhibit heavy reliance on wholesale funding, but liquidity risks appear contained because the average term structure of wholesale funding is sufficiently long. - LCR sensitivity re-parameterization (30-day horizon) outcomes: - Re-parameterization leads LCR ratios for some banks to decline below 100 percent but remain above 85 percent in all cases under the stressed parameters tested. - Trade-off highlighted: - Extending maturity of wholesale funding reduces roll-over risk but can lock-in higher interest costs, reducing resilience to interest-rate declines. ### Household sector — micro-data findings and DSGE housing model results - Household balance sheet and mortgage market statistics: - Household financial assets rose by 23 percent from €1,772 to €2,181 billion during 2010–2015. - Household financial liabilities increased by 6.3 percent during 2010–2015. - Households’ debt-to-net disposable income ratio: 294 percent in 2010 → 277 percent in 2015. - Debt-to-Total Assets at end-2014 = 25.7 percent. - 80 percent of total assets comprises home equities (34 percent), life insurance (5 percent) and pension entitlements (41 percent). - Mortgage market composition and risks: - Outstanding mortgages: peaked at €652 billion in 2012 → €639 billion by 2015. - Mortgage share of GDP: 97 percent → 94 percent during 2013–2015. - Share of underwater mortgages: one quarter of all mortgages at end-2015. - Underwater mortgages concentrated among young borrowers: about 40 percent age 20–30; 55 percent age 30–40. - Aggregate mortgage arrears: 2.3 percent in 2013Q1 → 1.6 percent at end-2015. - Mortgages with originating LTV > 100 percent and originating LTI ≈ 4 times: 20 percent of total mortgages at end-2015 (DNB loan-level dataset). - At end-2015: 37 percent of new mortgage loans have LTV ratios above 100 percent; 50 percent have LTI ratios above four times. - Interest-only (IO) mortgages: about 55 percent of total mortgages at end-2015; bulk of IO mortgages mature from 2030 onwards. - DNB (2016) estimate: 60 percent of outstanding IO mortgages will not be fully covered by contractual payments or pledged accounts at maturity. - Mortgages with pledged accounts: about 30 percent of total mortgage loans at end-2015. - Micro-simulation stress-test results (IPO / DNB loan-level): - Young, low-income households with high LTV and LTI ratios are most vulnerable. - At LTV over 120 percent: largest increase in share of high-risk borrowers; absolute level fairly low at 2.4 percent (figure from micro-simulations). - At LTI over six times: share increases by 60 basis points from 3.7 percent to 4.3 percent. - Borrowers under age 35 and age 35-45 more likely to face income shocks; over 65 nearly unaffected. - DNB loan-level data (end-2015): share of young borrowers (under 30) with originating LTV and LTI ratios above 100 percent and 5 times was 54 percent and 64 percent, respectively; analogous figures for age over 70 were 4 percent and 24 percent. - General-equilibrium (DSGE) housing model simulations: - Simulated a one-standard deviation increase in variance of housing returns; compared responses for LTV = 80 percent, 90 percent, 100 percent. - Findings: declines in output, employment, and consumption are small when LTV = 80 percent, markedly larger for LTV = 90 percent and LTV = 100 percent. Higher LTV → larger macroeconomic and default responses. - The model highlights that mortgage interest deductibility increases household indebtedness and that higher dispersion of housing returns raises default probabilities and credit spreads. ### Corporate sector — ORBIS-based analysis and sensitivity tests - Sample coverage and data caveats: - ORBIS sample: Total observations 78,833; Number of firms 17,215 (sample composition varies and presents survivorship bias and reporting lags). - Focus interpretation on variations over time rather than absolute levels. - Key corporate-sector metrics and concentration: - Outstanding corporate debt concentrated in large firms: by 2015, 92 percent of total debt concentrated among large firms. - Sectoral concentration (2015): >60 percent of corporate debt in manufacturing; 15 percent in trade. - Profitability, liquidity, leverage: - ROE recovered across firm sizes; domestic ROE 2015 = 16.4 percent (median). - Liquidity buffers maintained; median liquidity ratio (Total) 2009 = 2.1; 2015 = 1.9 (times). - Median debt-to-equity (Total) 2009 = 0.6; 2015 = 0.5. - Financial vulnerability (ICR) and sensitivity analysis: - Definition: ICR = EBITDA / interest payments; ICR < 1 signals financial vulnerability. - Share of firms with ICR < 1: 2009 = 27.0 percent; 2015 = 16.4 percent. - About 16.5 percent of firms exhibited ICRs below 1 in 2015 and these firms held about 21.5 percent of total corporate debt in 2015. - Sensitivity shock calibrations: - “Adverse” scenario: 80 basis points interest rate hike; 2.5 percent deflation; 9 percent nominal profit shortfall. - “Severe” scenario: 200 basis points interest rate increase; 6 percent deflation; 15 percent profit contraction. - Under the “severe” combined shock: - Median ICR for all firms falls to 6.5 (reported as 6.5). - Share of firms with ICR < 1 increases from 16.4 to 18.2 percent. - Proportion of debt held by firms with ICR < 1 increases from 21 to 26 percent. - Interest-rate increases dominate impact given the assumed calibrations. - Policy implication: address data gaps for more granular corporate balance-sheet investigation. ### Key recommendations (timeframes: I = within one year; NT = one–three years; MT = three–five years) - Risk analysis of the banking sector - Enhance DNB’s top-down solvency stress testing framework: - Develop satellite models of credit risk for multiple countries to account for large banks’ significant foreign activities. (NT) — DNB - Adjust DNB’s framework to exploit more granular EBA template data for net interest income calculation when stress tests occur at the European level. (NT) — DNB - Enhance DNB’s monitoring and stress testing framework for liquidity risk: - Refine framework to allow differentiated analysis by currency and ensure adequate and consistent reporting of cash flows (including derivatives and off-balance sheet positions) under liquidity monitoring metrics. (NT) — DNB - Household sector - Enforce an industry-wide standard approach to informing interest-only mortgage borrowers of their estimated repayment shortfalls. (NT) — DNB, AFM - Support tightening of LTV and DSTI limits to build borrower resilience (policy implication from micro and DSGE results). - Encourage industry measures to address IO mortgage risks: - Strengthen incentives to switch IO loans to annuity or linear mortgages; encourage voluntary prepayment and accumulation of financial assets; promote active use of equity release for older borrowers. - Develop standardized borrower information (example: UK biennial color-coded letters practice cited). - Corporate sector - Fill data gaps on the nonfinancial corporate sector to allow more granular investigation. (NT) — DNB, CBS *International Monetary Fund — Executive Summary (Netherlands) contained in the provided content unit.* ### EXECUTIVE SUMMARY ___________________________________________________________________________ 5 ### EXECUTIVE SUMMARY ### Macroeconomic context and key vulnerabilities - Gradual domestic recovery amid an uncertain international context. - Key domestic vulnerabilities: - High household debt and prevalent negative equity among young borrowers. - Substantial portion of households with a loan-to-value ratio above 100 percent. - Banks: improved capitalization since the crisis, but contracted balance sheets, low profitability, and significant reliance on wholesale funding. - Ongoing low interest rate environment and slow credit growth weigh on financial institutions. ### FSAP focus and stress-testing design - Tests examined overall risks and vulnerabilities with emphasis on three “triple threats”: - Low profitability from continued/further decline in interest rates and low credit growth. - High indebtedness of the household and nonfinancial corporate (NFC) sectors. - Banks’ reliance on wholesale funding and exposure to cross-border linkages. - Adverse (extreme but plausible) scenario characteristics: - Cumulative decline of GDP equal to two standard deviations relative to the baseline: 6.9 percentage points over two years and 8.7 percentage points over three years. - Downside external risks including Euro area recession and deflation, sovereign shocks, tighter global financial conditions. - Domestic shocks including a decline in house prices impacting private consumption and investment. - Combined shocks produce a V-shaped recession with negative growth of 3.7 percent in 2017. - Scenario incorporates strains on funding markets and pressure on interest margins. ### Banking sector — profitability and solvency results - Baseline projections: - Net interest income (NII) for the largest six banks is projected to decline from 1.3 percent of assets in 2015 to 1.1 percent of assets in 2017–18. - Overall profitability for the banking system is projected to fall from 0.50 percent of assets in 2015 to about 0.15 percent in 2018. - Adverse scenario (risk-weighted capital outcomes): - Fully loaded common equity Tier 1 (CET1) for the largest six banks declines from 13.5 percent in 2015 to 10.2 percent in 2018. - All banks remain above the minimum CET1 hurdle rate of 7 percent over 2016–18; no system capital shortfall on this metric. - Leverage outcomes and Tier 1 capital shortfall: - Fully loaded leverage ratio for the system (6 largest banks) declines from 3.8 percent to 3.2 percent in the adverse scenario. - One significant bank could fall just below the 3 percent minimum hurdle considered in the exercise. - Shortage of Tier 1 capital in the adverse scenario equivalent to €3.8 billion or 0.6 percent of GDP. ### Sensitivity and concentration analyses - Concentration risk (defaults of largest non-sovereign exposures): - When no haircut is applied to collateral values, large banks can withstand simultaneous defaults of up to four largest exposures without undercapitalization (relative to an 8.5 percent Tier 1 hurdle rate). - Under extreme assumptions with 50 percent haircuts on collateral values: - One of the largest banks exhibits a Tier 1 capital shortfall upon simultaneous default of the three largest exposures. - Other systemically important banks show undercapitalization with simultaneous default of the five largest exposures. - RWA-floor sensitivity for mortgage portfolios: - Current IRB mortgage risk weights average 15.4 percent for the largest banks (variation across banks). - Tests increase RWAs toward standardized portfolio risk-weight (35 percent) at specified levels: - RWAs increase to 21 percent (60 percent of 35 percent), 28 percent (80 percent of 35 percent), and 31.5 percent (90 percent of 35 percent). - Corresponding declines in CET1 ratios (based on end-2015 capital ratios) are equivalent to 0.7, 1.4, and 1.8 percentage points, respectively. ### Liquidity stress tests - Cash flow–based liquidity stress testing calibrated by liability type and asset haircuts. - Resulting withdrawals of funding equivalent to 15–20 percent of the initial stock within the first two to three months (variation across banks). - Maturity-ladder analysis results: - Except for a small institution, all banks could confront persistent and sizable withdrawals without emergency liquidity assistance (ELA) from the European Central Bank (ECB) for periods longer than six months. - System exhibits heavy reliance on wholesale funding, but liquidity risks appear contained due to sufficient average term structure of that funding. ### Household sector analysis and policy implications - Micro-level analysis (income and indebtedness) and a macroeconomic general equilibrium housing model were used. - Vulnerability findings: - Young, low-income households with high LTV and LTI ratios are the most vulnerable to adverse shocks. - Higher mortgage interest deductibility (MID) encourages higher indebtedness (higher LTV). - Households are more vulnerable to housing-outlook changes the higher the LTV ratio. - Model simulations: increase in variance of housing returns leads to declines in output, employment, and consumption; the decline is small if the LTV ratio is below 80 percent. - Higher LTV leads to earlier household default. - Policy implication: - Supports the financial stability role of LTV and debt-service-to-income (DSTI) limits; authorities are encouraged to tighten these measures further to build borrower resilience. ### Corporate sector analysis - Overall NFC recovery: - Nonfinancial corporates have restored profitability and strengthened equity-to-debt ratios overall. - Heterogeneity and vulnerabilities: - Pockets of vulnerabilities exist in certain segments. - Significant heterogeneity across firm categories suggests data gaps; more granular data needed for detailed corporate balance sheet investigation. ### Key recommendations (timeframes: I = within one year; NT = one–three years; MT = three–five years) - Risk analysis of the banking sector - Enhance DNB’s top-down solvency stress testing framework: - Develop satellite models of credit risk for multiple countries to account for large banks’ significant foreign activities. (NT) — DNB - Adjust DNB’s framework to exploit more granular EBA template data for net interest income calculation when stress tests occur at the European level. (NT) — DNB - Enhance DNB’s monitoring and stress testing framework for liquidity risk: - Refine framework to allow differentiated analysis by currency and ensure adequate and consistent reporting of cash flows (including derivatives and off-balance sheet positions) under liquidity monitoring metrics. (NT) — DNB - Household sector - Enforce an industry-wide standard approach to informing interest-only mortgage borrowers of their estimated repayment shortfalls. (NT) — DNB, AFM - Corporate sector - Fill data gaps on the nonfinancial corporate sector. (NT) — DNB, CBS _International Monetary Fund — Executive Summary (Netherlands) contained in the provided content unit._ ### 2. In addition, some features of the financial system and broader economy increase their ### cr1795 - 2. In addition, some features of the financial system and broader economy increase their ### Vulnerabilities in the financial system and broader economy - Household debt is high and negative equity among young borrowers is prevalent. A substantial portion of households have an LTV above 100 percent. - Bank balance sheets have contracted; profitability has only recently started to rebound from historically low levels. - Banks remain significantly reliant on wholesale funding. - Financial institutions face challenges from the continuing low interest rate environment. - The share of mortgages with fixed interest rates over five years is 73 percent. ### Triple threats to financial stability (FSAP focus) - Identified vulnerabilities and corresponding risks: - Low profitability due to a continuation/further decline in interest rates and low credit growth. - May reduce net interest income for banks. Low interest rates may further impact the financial positions of insurance companies and pension funds. - The indebtedness of the household and NFC sectors. - Increases probability of default of borrowers and exacerbates the impact of adverse shock to house and asset prices. - Banks’ reliance on wholesale funding and exposure to cross-border linkages. - Exacerbates the probability of an adverse funding shock and could negatively affect bank profitability. ### Objectives and role of FSAP stress tests - Assess the capacity of the banking system to withstand extreme but plausible macroeconomic shocks. - Explore weaknesses in the financial system and channels of transmission of adverse shocks. - Identify priorities for policy actions (e.g., reducing specific exposures, building capital and liquidity buffers). - Help authorities identify informational and methodological gaps and assess preparedness to deal with financial stress. - FSAP stress tests differ from central bank tests (e.g., DNB, ECB) in methodology; the FSAP used separate credit risk models and complementary methodologies. - Tests were carried out in close cooperation with the DNB and ECB with access to detailed supervisory data. ### Stress test coverage and methodologies - Main stress tests covered six banks that account for 87 percent of assets in the system: - ING Bank N.V.; Coöperatieve Rabobank U.A.; ABN AMRO Bank N.V.; Bank Nederlandse Gemeenten N.V.; Nederlandse Waterschapsbank N.V.; SNS Bank N.V. - Perimeter of analysis: consolidated group. - Scope broader than 2016 EBA Stress Test (EBA covered four banks representing 81 percent of assets). - Additional sensitivity tests included four DNB-supervised banks accounting for 3 percent of total assets: NIBC Bank N.V.; LeasePlan Corporation N.V.; Achmea Bank N.V.; F. van Lanschot Bankiers N.V. - Types of tests: - Solvency: Top-Down (TD) exercises based on macroeconomic scenarios and sensitivity analyses; TD complements but contrasts with Bottom-Up (BU) EBA approach. - Liquidity: TD liquidity tests using a maturity ladder (cash-flow analysis by maturity buckets) and supervisory information to assess capacity to confront large withdrawals. - Contagion: Domestic interbank exposures, interlinkages within domestic financial system, and cross-border exposures. - Household sector analysis: - Loan-level stress testing to approximate probability of default (PD) of mortgages under baseline and adverse scenarios, allowing segmentation by household characteristics (e.g., LTV ratios). - DSGE analysis of macrofinancial role of household indebtedness, evaluating relationship between LTV ratios (and potential LTV regulation adjustments) on household consumption volatility and growth. - NFC sector analysis: - Used macro- and micro-level data; assessed firms’ debt service capacity (e.g., interest coverage ratio) by firm size, sector, and ownership. - Balance-sheet sensitivity analyses calibrated to bank stress testing scenarios to estimate potential increase in share of financially distressed firms (interest cover ratio (ICR) less than one). - Used GFSR methodology with stressed ICR ratios to evaluate potential bank losses on NFC exposure. ### Interpretation caveats and limitations of stress tests - Stress tests are conceptually challenging and results must be interpreted with caution. - Use of macroeconomic and satellite models introduces estimation uncertainty because models are estimated on historical data. - Choices on severity of shocks and selection of historical episodes and data series affect outcomes. - Adverse scenarios represent combinations of external and domestic shocks with very low ex-ante probability; they should not be interpreted as macroeconomic “forecasts.” ### Key external and domestic risks highlighted - External risks: - Reversion in international integration and policy coordination in large economies could weigh on global growth and increase financial market volatility; post-Brexit uncertainty could weigh on confidence and investment. - Global economic and political fragmentation could sharply raise risk premia with flight to safety, triggering asset price drops, reduced insurer solvency and pension coverage ratios, reduced bank capital ratios, redemptions in investment funds, and higher wholesale funding costs. - Structurally weak growth in the Euro Area could reduce demand for Dutch exports, exacerbate NFC credit risks, and potentially trigger deflation—weakening banks via asset quality deterioration and weak credit demand. - Slowdown in China or tighter conditions in EMs could reduce global growth and affect the Netherlands through trade and financial channels given wide international exposure. - Domestic risks: - Reversal in recent recovery in house prices could weaken household balance sheets and dampen domestic demand. - Dutch banks’ high exposure to households and NFCs would transmit such shocks to banks’ asset quality and domestic consumption. ### Macroeconomic scenarios used in stress tests - Baseline scenario: based on IMF World Economic Outlook (WEO) projections as of August 2016; incorporates post-Brexit downward revisions to real GDP growth equivalent to 0.1 and 0.3 percentage points in 2016 and 2017, respectively. - Adverse IMF scenario: extreme but plausible, reflecting downside external risks and a domestic house price shock. - The IMF adverse macroeconomic scenario results in a cumulative decline of real GDP relative to the IMF WEO baseline equivalent to 2 standard deviations (6.9 percentage points, calculated based on data covering 1980–2015) over two years, and 1.9 standard deviations (or 8.7 percentage points) over three years. - A standard deviation of (two-year cumulative) real GDP growth, calculated based on data for the period 1980–2015, is equal to 3.4 percentage points. - The IMF adverse scenario exhibits a V shape (sizable decline in 2017) while the EBA scenario is U-shaped. - Stress-test horizon: three-year horizon (2016–18), based on data available through December 2015. ### Selected numeric paths in baseline and adverse scenarios (Table 2 excerpts) - Real GDP growth (in percent change) - Baseline: 2015 = 2.0; 2016 = 1.7; 2017 = 1.6; 2018 = 1.9 - Adverse IMF: 2015 = 2.0; 2016 = 0.1; 2017 = -3.7; 2018 = 0.2 - Short term interest rate (in percent) - Baseline: 2015 = 0.0; 2016 = - 0.3; 2017 = - 0.4; 2018 = - 0.2 - Adverse IMF: 2015 = 0.0; 2016 = 0.1; 2017 = 0.7; 2018 = 0.7 - Private consumption (in percent change) - Baseline: 2015 = 1.8; 2016 = 1.9; 2017 = 1.9; 2018 = 1.7 - Adverse IMF: 2015 = 1.8; 2016 = 0.1; 2017 = -4.4; 2018 = -1.0 - Long term interest rate (in percent) - Baseline: 2015 = 0.8; 2016 = 0.3; 2017 = 0.5; 2018 = 0.6 - Adverse IMF: 2015 = 0.8; 2016 = 0.8; 2017 = 1.6; 2018 = 1.7 - Private investment (in percent change) - Baseline: 2015 = 11.6; 2016 = 6.6; 2017 = 4.6; 2018 = 4.0 - Adverse IMF: 2015 = 11.6; 2016 = -0.5; 2017 = -16.9; 2018 = -4.2 - Mortgage interest rate (in percent) - Baseline: 2015 = 2.9; 2016 = 2.8; 2017 = 2.7; 2018 = 2.5 - Adverse IMF: 2015 = 2.9; 2016 = 3.0; 2017 = 3.2; 2018 = 3.0 - Consumer price inflation rate (in percent) - Baseline: 2015 = 0.2; 2016 = 0.2; 2017 = 0.9; 2018 = 1.0 - Adverse IMF: 2015 = 0.2; 2016 = 0.0; 2017 = -1.4; 2018 = -2.5 - Interest rate on corporate loans (in percent) - Baseline: 2015 = 1.7; 2016 = 1.7; 2017 = 1.6; 2018 = 1.5 - Adverse IMF: 2015 = 1.7; 2016 = 1.9; 2017 = 2.2; 2018 = 2.0 - Unemployment rate (in percent) - Baseline: 2015 = 6.9; 2016 = 6.6; 2017 = 6.4; 2018 = 6.1 - Adverse IMF: 2015 = 6.9; 2016 = 7.2; 2017 = 8.8; 2018 = 8.6 - Fiscal balance ratio (in percent) - Baseline: 2015 = -1.8; 2016 = -1.6; 2017 = -1.0; 2018 = -0.7 - Adverse IMF: 2015 = -1.8; 2016 = -1.7; 2017 = -1.4; 2018 = -1.5 *Source: IMF staff.* ### 17. Table 3 describes the different layers of numerical assumptions that underlie the ### cr1795 - 17. Table 3 describes the different layers of numerical assumptions that underlie the ### Adverse IMF scenario: common and country-specific layers - Common EA layer: - Driven by a tightening of global financial conditions accompanied by credit-cycle downturns in EM economies. - Abrupt decompression of risk asset premia amplified by low secondary market liquidity. - Within the EA, tensions particularly affect high spread economies; reflected in higher money market interest rates—and banks’ cost of short-term funding—and long-term government bond yields. - Monetary normalization in the systemic advanced economies fails to materialize in the stress testing period (2016–18). - Country-specific layer: - Housing prices decline by 15 percent, with adverse impact on domestic consumption and residential investment. - Heavy reliance on wholesale funding by Dutch banks makes them vulnerable to turmoil in international financial markets. - Under the country-specific layer, the cost of short-term funding rises by 75 basis points. - Combining both layers, banks’ cost of funding increases by 100 basis points. ### Funding-cost pass-through and interest-rate transmission - Differentiated impact across funding sources with a 50 percent pass-through of funding-cost shocks to lending rates. - Rationale: - Maturity transformation: loans have longer maturity/time-to-repricing than deposits, so shocks cannot be passed on immediately. - When repricing occurs, only 50 percent of funding shocks are passed to borrowers. - Differentiation across banks: those more reliant on wholesale funding face larger shocks; those relying on overnight/short-term deposits face smaller shocks. - Numerical examples and figures (as presented): - Funding cost shock: + 100 bps (average across all funding sources) - Loans: + 50 bps pass-through at loan repricing - Securities: + 0 bps in the illustrated example - All deposits: deposit-level shock factors (as listed): 0.84 overall; Overnight deposits 0.54; Redeemable at notice deposits 1.08; Time deposits 1.30; Debt securities 1.30. - Overall funding shock (deposits + securities) 0.99 (debt securities = 33% of deposit+securities funding) ### Table 3: Selected macro-financial shock assumptions (Common | Country-specific | Total) - Long term government bond yield; Duration risk premium shocks - High Spread Euro Area Economies: +200 basis points | +200 basis points - Low Spread Euro Area Economies (excl. Netherlands): +50 basis points | +50 basis points - Japan, United Kingdom, United States: +100 basis points | +100 basis points - Netherlands: +50 basis points | +75 basis points | +125 basis points - Real equity price; Equity risk premium shocks - China, Euro Area, Japan, United Kingdom, United States: −20 percent | −20 percent - Money market interest rate spread; Credit risk premium shocks - China, High Spread Euro Area Economies: +100 basis points | +50 basis points | +150 basis points - Low Spread Euro Area Economies (excl. Netherlands), Japan, United Kingdom, United States: +25 basis points | +50 basis points | +75 basis points - Netherlands: +25 basis points | +75 basis points | +100 basis points - Real bilateral exchange rate; Currency risk premium shocks - Euro Area: +5.0 percent | +5.0 percent - High Spread Euro Area Economies: +2.0 percentage point | +2.0 percentage point - Low Spread Euro Area Economies (excl. Netherlands): +1.0 percentage point | +1.0 percentage point - Netherlands: +1.0 percentage point | +1.0 percentage point - Loan default rate shocks - +0.3 to +4.7 percentage points | +0.3 to +4.7 percentage points - Private investment; Investment demand shocks - Common: −4.0 percent | Country-specific: −4.0 percent | Total: −4.0 percent (table shows variations for Germany and Netherlands separately—see below) - Private consumption; Consumption demand shocks - Common: −1.0 percent | Country-specific: −1.0 percent | Total: −1.0 percent - Germany (illustrative country-specific totals shown) - Private investment; Investment demand shocks: −4.0 percent | −8.0 percent | −12.0 percent - Private consumption; Consumption demand shocks: −1.0 percent | −2.0 percent | −3.0 percent - Netherlands (country-specific totals shown) - Private investment; Investment demand shocks: −4.0 percent | −16.0 percent | −20.0 percent - Private consumption; Consumption demand shocks: −1.0 percent | −2.0 percent | −3.0 percent ### Layered scenario timing (as labeled) - Layer 1: Tightening of financial conditions in systemic economies, 2016Q3 – 2017Q2 - Layer 2: Fiscal consolidation in the Euro Area, 2016Q3 – 2018Q2 - Layer 3: Credit cycle downturns in emerging market economies, 2016Q3 – 2018Q2 - Layer 4: Suppressed economic risk taking worldwide (excl. Germany and Netherlands), 2016Q3 – 2018Q2 ### Banks’ balance sheets, business models, and international exposures - Business-model summaries (largest banks, 2015 data): - ING, ABN Amro and Rabobank: about 70-75 percent of balance sheets in loans. - Loan book composition: mortgage loans to households 40–50 percent of the loan book; corporate loans 30–40 percent. - Funding: deposits 50–70 percent of total assets; securities 20–30 percent of total assets. - BNG and NWB: public, focus almost entirely on loans to the public sector; fund predominately with securities. - SNS: largest deposit base; focuses almost entirely on mortgage loans. - International exposures (2016 EU-wide stress test data): - ING: about 30 percent of IRB portfolio in the Netherlands; 70 percent to clients outside the Netherlands. - Five largest foreign counterparties: Germany 14.2 percent, Belgium 13 percent, United States 5.8 percent, Australia 4.6 percent, United Kingdom 4.6 percent; 5.4 percent invested in Spain and Italy. - Rabobank: about 68 percent of IRB portfolio in the Netherlands. - Five largest foreign counterparties: United States 11.6 percent, United Kingdom 3.4 percent, Australia 2.1 percent, Switzerland 2.3 percent, New Zealand 1.5 percent; 0.9 percent in Brazil. - ABN Amro: about 76 percent of IRB portfolio in the Netherlands. - Five largest foreign counterparties: France 3.4 percent, United States 3 percent, Germany 2.1 percent, Belgium 1.6 percent, United Kingdom 1.6 percent; holdings in Singapore 1.4 percent, Marshall Islands 0.9 percent, Bermuda 0.7 percent. ### Satellite models and methodologies for macrofinancial risk transmission - Credit losses: - PDs under the adverse scenario were estimated for 11 countries: Australia, Belgium, France, Germany, Italy, the Netherlands, Poland, Spain, Switzerland, the United Kingdom, and the United States. - A dynamic panel data econometric model estimated PDs; bank- and country-specific PDs determined by real GDP growth rates and real interest rates in a non-linear fashion. - Non-linear model implications: - Larger subsequent increases in PDs for exposures with higher starting PDs. - PDs rise at an increasing rate as macroeconomic conditions deteriorate further. - Cross-country variation in macroeconomic conditions and bank exposures affect overall PD increases. - PD multipliers and exposures: - System-wide point-in-time (PIT) PD multiplier in the adverse scenario: 2.4 times the starting level (overall), with variation across exposure types and countries. - Average multipliers (across all banks and countries): - Corporate credit exposures: 2.8 - Consumer credit exposures: 2.5 - Mortgage loans: 1.7 - Country variation example: multiplier values for corporate exposures range from 2.4 to 3 in low spread EA economies (Netherlands, Germany, Belgium) and from 5 to 6 in high spread EA economies. - Loss given default (LGD) adjustments: - Empirical association: when PDs rise from 1 to 2 percent, average recovery rates decline from 50 percent to 40 percent (LGDs rise from 50 percent to 60 percent). - PD multiplier equal to 2 associated with an LGD multiplier equivalent to 1.2. - LGDs adjusted according to formula shown: 0.6 x LGD Δ = PD Δ (as presented in source). ### Net interest income, repricing, and market losses - Net interest income projections: - Highly granular approach using bank-provided maturity structure and time-to-repricing data for individual banks. - Stress tests performed under the “constant balance sheet assumption”; maturing loans renewed at “new business” rates. - In the event of default, payments assumed suspended; interest proceeds adjusted by PDs in the “trees” calculation. - Market valuation losses on debt securities: - Measured via changes in yields leading to re-pricing using a modified duration approach. - Sovereign yield curves constructed by linear interpolation of short- and long-term interest rates as specified in the macro scenarios. - Valuation change formula (as presented): Valuation = − MD * MD y (where MD is modified duration and MD y is the change in yield matching the portfolio’s modified duration). ### Hurdle rates for solvency assessment - Regulatory frameworks applied: Basel III (fully loaded), EU and national frameworks as defined by CRD IV, Regulation on Prudential Requirements, national law and DNB regulations. - Hurdle rates for total capital, Tier 1 capital and Common Equity Tier 1 capital set according to current minimum requirements (Table 4). - Leverage ratio: Basel 3 percent leverage ratio used (no legal threshold). - Stress tests based on minimum capital ratios under Pillar 1; individual Pillar 2 requirements not taken into account. - Every ratio computed using Basel III fully loaded definitions. *Source: IMF staff and authorities, as presented in the IM F country chapter excerpt.* ### 30. Results show that low interest rates are expected to weigh down on bank profitability ### 30. Results show that low interest rates are expected to weigh down on bank profitability ### Impact of low interest rates on net interest income (Box 2) - Low interest rates and a prolonged flat yield curve are projected to cause significant deterioration in banks’ net interest income under the baseline scenario. - Lending rates are set to decline further as fixed income assets mature and are repriced, with negative effects on net interest income. - Net interest income for the largest six banks: - €29.4 billion (1.3 percent of assets) in 2015 - €25 billion in 2017 (about 1.1 percent of assets) - €24 billion in 2018 (about 1.1 percent of assets) - Overall banking-system profitability under the baseline: - 0.50 percent of assets in 2015 - About 0.15 percent of assets in 2018 - Main driver: dominant effect of declining net interest income, partly mitigated by lower credit losses and taxes - Mechanisms described: - Banks hedge fixed-to-floating risk via swaps at origination; hedging reduces interest rate risk because liabilities have shorter maturities than loans. - When the yield curve flattens and term spreads narrow, maturity transformation becomes less profitable: swaps repriced at rollover reduce the term spread. - Additional downward pressure when borrowing rates hit the zero lower bound for some liabilities. - Impact on net interest income is lagged because loan portfolios roll over gradually; new business loans exhibit significantly lower interest rates than outstanding loans. ### Solvency stress-test results (baseline and adverse scenarios) - Stress tests indicate banks are affected significantly by extreme but plausible adverse macroeconomic conditions, yet maintain capital ratios above regulatory minimums. - Common Equity Tier 1 capital ratio (CET1) for the largest six banks: - 13.5 percent in 2015 - Declines to 10.2 percent in 2018 under adverse scenario - Regulatory hurdle and capital shortfall: - Minimum hurdle CET1 ratio: 7 percent over 2016–18 - All banks’ CET1 ratios remain above the 7 percent hurdle; system does not have any capital shortfall on CET1 metric - Leverage ratio outcomes: - System (six largest banks) leverage ratio declines from 3.8 percent to 3.2 percent in the adverse scenario - One of the largest banks would fall below the minimum 3 percent leverage-ratio hurdle - Result implies a Tier 1 capital shortage in the adverse scenario equivalent to €3.8 billion or 0.6 percent of GDP - Drivers of CET1 decline in adverse scenario: - Deterioration in net interest income, credit and security losses, and higher RWAs ### Adverse scenario: profitability and loss decomposition - Overall profitability under adverse scenario: - Declines from 0.5 percent of assets in 2015 to -0.3 percent in 2017–18 - Net interest income declines from 1.3 percent of assets in 2015 to 0.8 percent of assets in 2018 - Credit loss impairments: - Rise from 0.2 percent of assets in 2015 to 0.5 percent of assets in 2015 (text contains this phrasing) - Credit losses decline to 0.3 percent in 2018 as the economy starts recovering - Securities valuation impact: - Net gain of 0.2 percent of assets in 2015 turns into a 0.2 percent loss in 2017 due to sharp increases in sovereign bond yields in the adverse scenario - Taxes and pre-tax losses: - Banks post losses on a pre-tax basis; reduced tax payments partially mitigate bottom-line decline - Credit-risk dynamics: - PIT PDs rise sharply in adverse scenario, reaching a peak equivalent to 2.4 times the initial level (system-wide), with significant variation by exposure type and country - Average multipliers (across all banks and countries) for PD increases: - Corporate exposures: 2.8 - Consumer credit exposures: 2.5 - Mortgage loans: 1.7 - Multipliers vary by country; corporate exposure multipliers range from 2.4 to 3 in low-spread EA economies and from five to six in high-spread EA economies - Market-risk losses on securities: - Duration-adjusted haircuts applied to securities in trading, available-for-sale, and fair-valued books cause losses equivalent to 2 percent of initial valuations - Mitigating factors: presence of floating-rate securities (not stressed), large allocations to low-spread countries, and short/intermediate portfolio durations ### Sensitivity tests: concentration risk and mortgage RWA floors - Concentration risk (default of largest exposures, excluding sovereign exposures): - Using supervisory data and varying recovery-rate assumptions - When no haircut applied to collateral values: - Large banks could withstand simultaneous defaults of up to four largest exposures without undercapitalization relative to an 8.5 percent Tier 1 hurdle rate - One small bank undercapitalized upon default of single largest exposure; one additional small bank undercapitalized upon simultaneous default of three largest exposures - With 50 percent haircuts on collateral values (extreme scenario): - One of the largest banks exhibits a Tier 1 shortfall upon simultaneous default of the three largest exposures - Other systemically important banks exhibit undercapitalization upon simultaneous default of five largest exposures - RWA floors for mortgage portfolios (sensitivity to increasing IRB-based mortgage RWAs from current levels): - Current average mortgage RWA (largest banks): 15.4 percent on average for the largest banks in the system (end-2015) - Floors tested relative to standardized portfolio 35 percent level; alternative floors raise IRB mortgage RWAs to: - 21 percent (low mortgage RWA floor) - 28 percent (medium mortgage RWA floor) - 31.5 percent (high mortgage RWA floor) - Impact on CET1 at end-2015 ratios: - Introduction of floors would cause declines in CET1 equivalent to 0.7, 1.4, and 1.8 percentage points of CET1 respectively - Note: tests are static and do not consider phase-in or dynamic behavioral responses or eligibility criteria for preferential mortgage treatment ### Liquidity stress tests and maturity-ladder analysis - Methodology: - Cash-flow based liquidity stress tests using a TD approach with contractual cash-flow time structure - Tests assess: (i) liquidity needs from outflows; (ii) available standby liquidity from inflows; (iii) buffers to offset liquidity gaps - Run-off rate calibration principles: - Wholesale funding run-off rates higher than retail; secured funding run-off rates lower than unsecured - Sight deposits and insured deposits assumed more stable than time deposits and uninsured deposits - Cash outflows assumed protracted—lasting up to two years—with run-off rates declining over time - Highest run-off rates applied to 0–30 day bucket; medium rates to 30–180 days; lowest to >180 days - Cash-inflow conversion (maturing loans to cash proceeds): - 40–50 percent rates applied to performing loans to retail and nonfinancial corporate customers - 100 percent rates applied to maturing loans to other entities (including financial institutions) and cash flows from debt securities - Banks’ liquidity buffers and central-bank access: - Banks could use sovereign and non-sovereign securities as collateral to obtain liquidity through ECB standard facilities - Emergency Liquidity Assistance (ELA) from the ECB allowed as last resort; requiring ELA to continue operating constitutes a test failure - Test outcomes: - Combined maturities and run-off rates result in withdrawals equivalent to 15-20 percent of the initial stock within the first two to three months (variation across banks) - Cumulative outflows and asset-sale needs vary by bank; banks with extended liability maturities face less intense outflows *Source: IMF staff (text extracted from the IMF country report).* ### 42. The results of the liquidity stress tests reveal that banks could handle significant ### cr1795 - 42. The results of the liquidity stress tests reveal that banks could handle significant ### Liquidity Stress Test Findings - All banks could confront persistent and sizable withdrawals of funding without ELA from the ECB for periods longer than six months. - Figure 15: Netherlands: Bank Liquidity Stress Tests — Cumulative Inflows, Outflows, and Use of Counter-Balancing Capacity (in percent of outstanding non-equity liabilities). - Cumulative flows reported by time buckets: Cummulative inflows; Cummulative outflows; Cummulative use of CBC (counter-balancing capacity). ### Wholesale Funding and Term Structure - Despite banks’ heavy reliance on wholesale funding, liquidity risks appear contained because the term structure of this type of funding has sufficient average length. - Wholesale funding dependence implications: - Extending the maturity of wholesale funding reduces roll-over (funding liquidity) risks. - Extending maturity can lock-in interest rate costs (when funds are raised at fixed rates), reducing the ability to lower costs when interest rates decline, as envisaged in the baseline scenario. ### Liquidity Coverage Ratio (LCR) Sensitivity Analysis - All banks currently have LCR ratios that are comfortably above the 100 percent requirement. - Re-parameterization of run-off rates (30-day horizon) assessed effects of larger net liquidity outflows: - Stable retail deposits: increased from 5 to 10 percent. - Deposits subject to higher outflows: increased from 12.5 to 20 percent (category 1) and from 17.5 to 25 percent (category 2). - Wholesale deposits: - Operational deposits covered by deposit guarantees: increased from 5 to 10 percent. - Wholesale deposits unprotected by guarantees: increased from 25 to 40 percent. - Result of re-parameterization: LCR ratios for some banks (including some large ones) would decline below 100 percent but in all cases would remain above 85 percent. - Footnote: Most non-deposit funding tends to generate cash outflows at longer than 30-day horizons, so applying higher run-off rates to these funding sources would not alter LCR calculations in a significant manner. ### Relation between Solvency and Liquidity Stress Test Results - Key remarks: - Is wholesale funding a “threat”? The most significant risk from dependence on wholesale funding is revealed in solvency tests: banks’ cost of funding could rise sharply under adverse conditions, adversely impacting net interest income and profitability. - Sudden withdrawal of wholesale funding would not pose an imminent risk, given that banks have been extending the maturity of this funding. - Trade-off: strengthening resilience to changes in interest rates versus improving resilience to funding liquidity shocks. ### Household Sector — Overview and Vulnerabilities - The housing boom-bust cycle and a double-dip recession left a declining, but still highly indebted household sector in the Netherlands. - Recent trajectories: - Household financial assets increased by 23 percent from €1,772 to €2,181 billion during 2010–2015. - Household financial liabilities increased by 6.3 percent during 2010–2015. - Households’ debt-to-net disposable income ratio decreased from 294 percent in 2010 to 277 percent in 2015. - Household debt-to-disposable income ratio in international comparison: among the highest in Europe (Figure 16). - Net wealth composition and liquidity: - Debt-to-Total Assets stood at 25.7 percent at end-2014. - 80 percent of total assets comprises home equities (34 percent), life insurance (5 percent) and pension entitlements (41 percent) — largely illiquid. - Considering only financial assets, households had about minus €170 billion net worth at end-2014. ### Mortgage Market Structure and Risks - Outstanding mortgage balances and market shares: - Mortgages peaked at €652 billion in 2012 and decreased to €639 billion by 2015. - Mortgage share of GDP declined from 97 percent to 94 percent during 2013–2015. - Securitization via SPVs declined; insurance companies increased market share from 3.8 percent to 7.7 percent. - Negative equity and arrears: - Share of underwater mortgages accounted for a quarter of all mortgages at end-2015. - Residential property prices fell by 21 percent during the double-dip recession. - Underwater mortgages are concentrated among young borrowers: about 40 percent among borrowers age 20–30; 55 percent among those age 30–40. - Aggregate mortgage arrears: 2.3 percent of total mortgages in 2013Q1; 1.6 percent at end-2015. - High LTV and LTI exposure: - Mortgages with originating LTV above 100 percent and originating LTI about four times account for 20 percent of total mortgages at end-2015 (DNB loan-level dataset). - At end-2015, 37 percent of new mortgage loans have LTV ratios above 100 percent; 50 percent have LTI ratios above four times. - Mortgages with higher originating LTV and LTI ratios have higher arrears rates (Table 5): example arrears shares at end-2015 include 0.4 percent for LTV < 60 percent and LTI 0-2, and 4.9 percent for LTV > 120 percent and LTI > 6. - Interest-only (IO) mortgages: - IO mortgages comprise about 55 percent of total mortgages at end-2015. - The bulk of IO mortgages will start to mature from 2030 onwards. - One quarter of mortgage borrowers fully rely on an IO mortgage; about 60 percent of IO borrowers combine an IO with an amortizing mortgage. - DNB (2016) estimate: 60 percent of outstanding IO mortgages will not be fully covered by contractual payments or pledged accounts at maturity. - Mortgages with pledged accounts account for about 30 percent of total mortgage loans at end-2015. ### Household Micro-Simulation and Stress Testing - Purpose: Assess impact of FSAP adverse scenario and heterogeneity across household segments using micro-level data. - Data and methodology: - Simulations conducted by DNB staff using internal mortgage credit risk model for top-down stress tests due to confidentiality of the Income Panel Survey (IPO) data. - Scenario inputs affect household income, mortgage characteristics, and each household’s DSTI ratio. - IPO dataset refined sample covers about 4.5 million units after exclusions (early retirees, students, those not in the labor force). - 2010 data wave used as starting point; DNB loan-level data used as supplementary information. - IPO sample characteristics: - 97.5 percent of mortgage loans in sample are outside Amsterdam. - Most common age categories of the oldest in households: 35-45 and 45-55 years old. - About 50 percent of loans have LTV ratio below 90 percent, with about 40 percent being in negative equity in the sample (slightly above the 2015 level). - Borrowers with LTI ratio under four times represent about 86 percent of the sample; 6 percent have LTI ratio above five times. *Source: IMF staff and De Nederlandsche Bank (from the cited PDF chapter).* ### 59. In addition to the micro-study, staff used a general equilibrium model with a housing ### cr1795 - 59. In addition to the micro-study, staff used a general equilibrium model with a housing ### General-equilibrium DSGE model with housing: setup and objectives - Purpose: simulate how uncertainty about house prices and household incomes affects macroeconomic conditions for different LTV ratios (80 percent, 90 percent, and 100 percent). - Calibrated to characteristics of the Dutch economy. - Simulated impacts on: - Variables governing home purchase decisions: mortgage rates (the spread above the benchmark rate), mortgage lending, and residential investment. - Macroeconomic variables from general equilibrium dynamics: output, employment, and consumption. - Model features: - Differentiates between households that serve mortgages and households that default. - Simulates changes in the distribution of defaults depending on financial and macroeconomic conditions. - Emphasizes high-risk households on the verge of default. - Households finance housing purchase ݌ ௛,௧ ݄ ௧ with mortgage loan ݈ ௧ subject to an LTV constraint. - Next period house return ܴ ௛,௧ାଵ equals rental rate plus house value increase; housing returns vary across households due to idiosyncratic shock ߴ ௜,௧ାଵ . - Banks charge a spread on mortgage loans above the deposit rate to cover expected credit losses; they set lending rates so expected return on mortgage portfolio ≥ funding cost (deposit rate ܴ ௗ,௧ ). ### Key model mechanisms and theoretical insights - Tax-deductibility of mortgage payments increases household indebtedness (Dutch tax laws: ߬ ௛ ൐0). - Banks incur a loss on default ( ߤ ௛ ൐0). - If households could freely choose indebtedness (LTV), they would increase indebtedness until expected default costs offset expected tax-deduction benefits; higher tax rates → higher indebtedness. - Higher dispersion (variance) of housing returns ߪ ௛,௧ increases default probability, raising banks’ expected default costs, prompting higher credit spreads and lending rates, reducing credit demand, residential investment, output, and employment. - Consumption declines only gradually relative to output/employment due to consumption habits and increased household liquidity as residential investment falls. ### Simulated quantitative responses (model results summary) - A one-standard deviation increase in the variance ߪ ௛,௧ of housing returns was simulated; responses shown for output, employment, and consumption. - Effects conditional on LTV: - LTV 80 percent: decline in output, employment and consumption is small. - LTV 90 percent and LTV 100 percent: responses of output, employment and consumption are significantly larger than for LTV 80 percent. - Comparative ordering preserved: higher LTV → larger macroeconomic and default responses. ### Table 6 — Composition of Residential Mortgages (IPO 2013 data, about 55,000 households; in percent of the sample) - Location - Amsterdam: 2.5 - Non-Amsterdam: 97.5 - Age of the oldest in a household - Age Under 35: 13.5 - Age 35-45: 29.3 - Age 45-55: 30.0 - Age 55-65: 15.5 - Age 65-75: 8.8 - Age 75+: 3.0 - Gross Income (threshold in euros) - Income Under 20 percentile 20 (49,220) - Income 20-40 percentile 20 (66,198) - Income 40-60 percentile 20 (83,560) - Income 60-80 percentile 20 (109,101) - Income 80-100 percentile 20 (277,522) - Current LTV Ratio - LTV Under 60: 29.4 - LTV 60-70: 6.2 - LTV 70-80: 6.5 - LTV 80-90: 6.8 - LTV 90-100: 7.2 - LTV 100-110: 8.4 - LTV 110-120: 9.8 - LTV 120+: 25.8 - Current LTI Ratio - LTI 0-2 times: 36.0 - LTI 2-3 times: 29.5 - LTI 3-4 times: 20.8 - LTI 4-5 times: 7.6 - LTI 5-6 times: 2.7 - LTI over 6 times: 3.4 - Source: DNB. ### Micro-simulation and distributional results (high-risk household shares) - Young, low-income households with high LTV and LTI ratios are most severely affected by adverse shocks. - Specific quantitative findings: - At LTV over 120 percent, the largest increase in the share of high risk borrowers occurs, even if the absolute level is fairly low at 2.4 percent given historically low default rates (upper left chart in Figure 17). - At LTI over six times, the share increases by 60 basis points from 3.7 percent to 4.3 percent (upper right chart in Figure 17). - Forward-looking vulnerabilities decrease with age: those under age 35 and age 35-45 are more likely to face an income shock; those over 65 are nearly unaffected (bottom left chart in Figure 17). - Based on DNB loan-level data at end-2015: share of young borrowers (under 30 years old) with originating LTV and LTI ratios above 100 percent and 5 times was 54 percent and 64 percent, respectively; analogous figures in age over 70 group were 4 percent and 24 percent. - Income groups: those under 20 percentile suffer most with share of high risk households increasing by about 60 basis points; those in 80-100 percentile are nearly unaffected due to ample liquidity buffer (bottom right chart in Figure 17). - Borrowers outside Amsterdam are more vulnerable than those in Amsterdam; on average, borrowers in Amsterdam earn higher income, providing buffers against shocks. ### Results synthesis - Higher LTV ratios increase household vulnerability to adverse changes in house prices and incomes. - Higher mortgage payment deductibility and higher tax rates increase household indebtedness. - Combined macro and micro evidence: borrowers with high LTV and LTI ratios are relatively more vulnerable to adverse shocks. ### Policy implications and recommendations - Support for using limits on LTV and DSTI (DSTI often called DSTI or DSTI caps in the text) ratios: - Authorities should further tighten measures aimed at building borrower resilience against future macrofinancial shocks. - IMF (2014) evidence: LTV limits and DSTI limits complement each other in reducing probabilities of defaults for borrowers and loss given defaults for lenders. - LTV limits without DSTI limits could leave borrowers’ capacity to service mortgages vulnerable to income shocks. - DSTI caps without LTV measures could leave lenders highly exposed to severe house price shocks. - Combining LTV and DSTI measures enhances resilience of households and lenders’ balance sheets. - Address risks from large share of interest-only (IO) loans: - Industry and authorities should act now to help borrowers with non-amortizing loans prepare for repayment by: - Strengthening incentives to switch loan types to annuity or linear mortgages. - Encouraging voluntary prepayment. - Encouraging accumulation of financial assets. - Promoting active use of equity release (especially for older borrowers). - Authorities should work with industry to develop a standardized approach to informing IO mortgage holders of their financial status vis-à-vis these loans and advise on early remedial options. - Example cited: United Kingdom practice—insurers send biennial color-coded letters since 2000 to endowment policyholders (red/amber/green) indicating likelihood of policy repaying mortgage principal and setting out options. *Source: This content unit is drawn from the IMF country report describing staff’s DSGE housing model simulations, loan-level microdata analysis, and policy conclusions.* ### 72. While comprehensive, the ORBIS database is not exempt of limitations that should be ### cr1795 - 72. While comprehensive, the ORBIS database is not exempt of limitations that should be ### Data limitations and interpretation guidance - ORBIS sample composition varies over time, limiting intertemporal analysis. - Survivorship bias likely overstates profitability indicators (panel includes surviving firms; bankrupt ones drop out). - Inclusion of newly-created firms (especially among SMEs) may boost indebtedness ratios because these firms have not broken even yet. - Number of observations drops markedly for year 2015 due to reporting lags, undermining accuracy of forward-looking analyses in fast-changing environments. - Observations decrease when indicators are computed by firm categories, creating data restrictions. - Guidance: focus interpretation on variations over time rather than absolute levels. ### Sample coverage by firm size (Average Sample Coverage by Firm Size) - Domestic: Number of observations 74,004; Number of firms 16,237 - Small (domestic): Number of observations 67,789; Number of firms 15,722 - Medium (domestic): Number of observations 4,226; Number of firms 1,122 - Large (domestic): Number of observations 1,990; Number of firms 535 - Foreign: Number of observations 4,829; Number of firms 978 - Small (foreign): Number of observations 2,951; Number of firms 801 - Medium (foreign): Number of observations 957; Number of firms 241 - Large (foreign): Number of observations 921; Number of firms 231 - Total: Number of observations 78,833; Number of firms 17,215 - Note: Small (1-49 employees), medium (50-249 employees), large (>249 employees). - Note 1/: The sum of small, medium, and large firms tend to exceed their reported number due to some firms switching categories over the years. ### Sample coverage by firm sectors (percent) - Agriculture: Domestic 0.8; Foreign 1.7; Total 0.9; Share of observations Domestic 0.6; Foreign 1.3; Total 0.6 - Manufacturing: Domestic 4.1; Foreign 16.5; Total 5.0; Share of observations Domestic 2.9; Foreign 11.5; Total 3.4 - Trade: Domestic 7.1; Foreign 25.8; Total 8.5; Share of observations Domestic 4.5; Foreign 16.8; Total 5.2 - Information: Domestic 2.4; Foreign 5.4; Total 2.6; Share of observations Domestic 1.4; Foreign 3.5; Total 1.6 - Finance: Domestic 56.6; Foreign 33.4; Total 55.0; Share of observations Domestic 66.5; Foreign 51.3; Total 65.6 - Housing: Domestic 3.7; Foreign 1.3; Total 3.5; Share of observations Domestic 3.2; Foreign 1.7; Total 3.1 - Services: Domestic 22.3; Foreign 14.6; Total 21.7; Share of observations Domestic 19.0; Foreign 13.0; Total 18.6 - Other: Domestic 3.0; Foreign 1.3; Total 2.9; Share of observations Domestic 1.9; Foreign 0.9; Total 1.9 - Total: Domestic 100; Foreign 100; Total 100; Share of observations Domestic 100; Foreign 100; Total 100 ### Profitability, liquidity, and leverage — key findings - ROE recovery: - Return on equity (ROE) ratios rebounded across all firm size categories after the 2011–2012 slump. - Profit especially high for small and medium-sized domestic firms. - Relative profitability lagging in housing and services; downward trend in manufacturing and trade sectors. - Liquidity: - Liquidity buffers continuously maintained, especially for small firms and in the finance and services sectors. - Debt-to-equity: - Debt-to-equity ratios remained stable for all firm sizes and most sectors. - Increasing leverage noted in manufacturing and trade sectors recently. - Housing industry deleveraging: debt-to-equity ratios brought down below 1 along the recovery that started in 2013. ### Concentration of outstanding corporate sector debt - Outstanding corporate sector debt concentrated in large firms and in the manufacturing sector. - Historical shift: - 20 percent of total corporate debt was held by SMEs in the aftermath of the financial crisis. - By 2015, 92 percent of total debt was concentrated among large firms. - Sectoral concentration in 2015: - More than 60 percent of corporate debt had been incurred in the manufacturing sector. - 15 percent in the trade sector. ### Financial vulnerability and Interest Cover Ratios (ICR) - Definition: - ICR = EBITDA / interest payments due on liabilities for the same time period. - Firms may be deemed financially vulnerable if ICR < 1 (with caveats about other resources or new liabilities). - Data notes: - EBITDA computed using ORBIS profit and loss account information. - Interest payments estimated by applying average corporate sector interest rates provided by the DNB for short-term (<1 year) and long-term (>1 year) interest rates to interest-generating components of liabilities. - Outstanding debt by firm size (percent of total), selected years: - 2009 Total 100.0; Small 10.3; Medium 12.2; Large 77.3 - 2011 Total 100.0; Small 15.5; Medium 15.1; Large 69.5 - 2013 Total 100.0; Small 6.3; Medium 15.2; Large 78.5 - 2015 Total 100.0; Small 1.4; Medium 6.4; Large 92.3 - Interest cover ratios (median, in percent) by firm size, selected years: - Domestic firms ROE: 2009 13.4; 2011 12.9; 2013 11.0; 2015 16.4 - Small ROE: 2009 11.8; 2011 11.7; 2013 10.8; 2015 16.5 - Medium ROE: 2009 14.5; 2011 13.6; 2013 10.6; 2015 17.9 - Large ROE: 2009 13.4; 2011 13.7; 2013 12.2; 2015 14.4 - Liquidity ratio (median, times) 1/ by firm size: - Total: 2009 2.1; 2011 2.2; 2013 2.2; 2015 1.9 - (1/ (Current Assets - Stocks) / Current Liabilities) - Liquidity ratio (median, times) by sector (domestic), selected years: - Total: 2009 2.1; 2011 2.2; 2013 2.2; 2015 1.9 - Finance: 2009 2.4; 2011 2.6; 2013 2.7; 2015 2.6 - Services: 2009 2.4; 2011 2.5; 2013 2.5; 2015 2.3 - Debt-to-equity ratio (median, times) by firm size: - Total: 2009 0.6; 2011 0.5; 2013 0.5; 2015 0.5 - Domestic small: 2009 0.7; 2011 0.8; 2013 0.7; 2015 0.7 - Domestic medium: 2009 0.6; 2011 0.6; 2013 0.5; 2015 0.4 - Domestic large: 2009 0.6; 2011 0.5; 2013 0.5; 2015 0.5 - Debt-to-equity ratio (median, times) by sector (domestic), selected years: - Total: 2009 0.6; 2011 0.5; 2013 0.5; 2015 0.5 - Housing: 2009 1.8; 2011 1.9; 2013 1.1; 2015 0.5 - Manufacturing: 2009 0.5; 2011 0.4; 2013 0.4; 2015 1.2 ### Distribution of ICRs and debt at risk - ICR distribution (share of domestic firms, percent of total): - ICR Total 2009/11/13/15 = 100.0 - ICR < 1: 2009 27.0; 2011 25.9; 2013 26.6; 2015 16.4 - 1 ≤ ICR < 2: 2009 3.7; 2011 3.9; 2013 3.1; 2015 4.0 - 2 ≤ ICR < 3: 2009 3.2; 2011 3.8; 2013 3.4; 2015 3.5 - ICR ≥ 3: 2009 66.1; 2011 66.3; 2013 66.9; 2015 76.1 - ICR of domestic non-financial corporations by sector (median, percent), selected years: - Total ROE: 2009 12.6; 2011 12.6; 2013 10.6; 2015 15.3 - Manufacturing ROE: 2009 17.3; 2011 17.8; 2013 14.4; 2015 14.3 - Services ROE: 2009 16.2; 2011 14.4; 2013 12.8; 2015 18.8 - Debt at risk (percent of each size category) — share of firms with low ICRs and their share of debt (selected entries, percent): - Total ICR < 2: 2009 65.3; 2011 57.1; 2013 27.3; 2015 19.7 - Total ICR < 1: 2009 49.7; 2011 34.9; 2013 34.9; 2015 31.4 - Small ICR < 2: 2009 55.8; 2011 43.0; 2013 15.9; 2015 12.6 - Small ICR < 1: 2009 55.0; 2011 42.6; 2013 34.1; 2015 26.2 - Medium ICR < 2: 2009 77.2; 2011 52.0; 2013 53.1; 2015 33.1 - Medium ICR < 1: 2009 66.2; 2011 58.1; 2013 44.3; 2015 34.8 - Large ICR < 2: 2009 64.7; 2011 59.8; 2013 24.2; 2015 18.4 - Large ICR < 1: 2009 46.1; 2011 29.7; 2013 30.5; 2015 20.3 ### Vulnerable segment and debt shares (2015) - About 16.5 percent of firms exhibited ICRs below 1 in 2015. - These firms held about 21.5 percent of total corporate debt in 2015. - Higher proportions of firms with ICR < 1 reported among small- and medium-sized categories. - Caveat: sample distortions likely overestimate the share of debt held by firms with low ICRs. - Sectoral concentration of debt held by vulnerable firms (limited observations): mostly in services and, to a lesser extent, in the housing industry. - Noteworthy trend: sharp decrease in the share of debt held by vulnerable firms from the peak during the 2012 recession across all firm sizes. ### Sensitivity analysis methodology - ICR recalculated using adjusted EBITDA and interest payments under three shock types: interest rate shock, profit shock, and combined shock. - Interest rate shock assumptions: - Uniform tightening of nominal effective interest rates along the yield curve. - Reduction in nominal profits to capture deflationary trends. - 70 percent of long-term debt assumed renegotiated at time environment deteriorates. - Gains on financial assets mitigate impact with a pass-through of 50 percent of the interest rate increase. - Profit shock assumption: - EBITDA drops in nominal terms (incorporates deflationary movements). - Combined shock: both interest rate and profit shocks applied together using the formulas described in the source. - Two scenarios: - “Adverse” scenario: 80 basis points interest rate hike; 2.5 percent deflation rate; 9 percent shortfall in nominal profits. - “Severe” scenario: 200 basis points interest rate increase; 6 percent deflation rate; 15 percent profit contraction. ### Sensitivity analysis results — key outcomes (Table 8 summary) - Dutch NFC sector broadly resilient to adverse macroeconomic shocks due to strong initial buffers. - Under the “severe” combined shock: - The median ICR for all firms falls to 6.5 percent (reported as 6.5). - Share of firms with ICR < 1 increases from 16.4 to 18.2 percent. - Proportion of debt held by firms with ICR < 1 increases from 21 to 26 percent. - Impact assessment: - Interest rate increases appear to dominate the impact relative to profit shortfalls, given the assumed calibrations. - Notable deviations in ICRs compared to baseline, especially for the largest enterprises. - Selected sensitivity outputs (ICR of the median firm, ICR < 2, ICR < 1; percent entries reported in source table): - Baseline median ICR (All firms) reported alongside scenario outcomes in Table 8 (values preserved in the original table). - Interest rate shock, profit shock, and combined shock show varying impacts across All firms, Small, Medium, and Large categories (see source table for exact percent entries). ### Conclusion and policy implication - Overall assessment: - Dutch NFC sector has emerged from the crisis with strong financial buffers, restored profitability, strengthened equity, and built-up liquidity buffers—preserving capacity to service debt in the face of shocks. - Some heterogeneity across firm categories indicates likely financial vulnerabilities in subcategories. - Policy recommendation from prudential viewpoint: - Swiftly address data gaps that hamper more granular investigation into corporate balance sheets. - Further investigation using a more granular and complete dataset is warranted given remaining pockets of vulnerabilities. *Source: ORBIS; IMF staff calculations; extracted from cr1795 (Netherlands country report).* ### 4. Risks and Buffers ### 4. Risks and Buffers ### Risks/factors assessed - Based on EBA 2016 EU-wide stress testing methodology. - Credit losses by exposure type and country. - Losses from debt instruments (sovereign and other issuers) in the banking and trading books. - Impact of funding cost shocks on net interest income. - Market risk, including foreign exchange risk. ### Behavioral adjustments - Static balance sheet assumption. - Payout rates consistent with past experience. - Dividends can only be paid out by banks that remain adequately capitalized. Payout rates consistent with past experience. ### Key assumptions table (domain-level) - Domain / Assumption / Bottom-Up by Banks (EBA) / Top-Down by Authorities / Top-Down by FSAP Team - Reporting Format for Results: Capital shortfall system wide; dispersion of capital ratios: min., avg., max.; percentage of assets that fail; for each hurdle rate (or range), share in whole system by asset. *Italic: Source: cr1795 - 4. Risks and Buffers* --- ### 5. Regulatory and Market-Based Standards and Parameters ### Calibration of risk parameters - PDs and LGDs: point in time for credit losses and through the cycle for stressed RWA calculations. (repeated for multiple exposure categories as in source) ### Regulatory/Accounting and Market-Based Standards - Hurdle rate: not specified. - Capital metrics: transitional and fully loaded Basel III regulatory requirements. - CET1, T1, CAR, leverage ratio. - Hurdle rates: as indicated in Table 4. - Capital metrics: fully loaded Basel III regulatory requirements. - CET1, T1, and leverage ratio. *Italic: Source: cr1795 - 4. Risks and Buffers* --- ### Banking Sector: Liquidity Risk (joint between authorities and FSAP team) ### Institutional perimeter - Institutions included (Top-Down by Authorities): The following seven banks: ING Bank N.V.; Coöperatieve Rabobank U.A.; ABN AMRO Bank N.V.; N.V. Bank Nederlandse Gemeenten; Nederlandse Waterschapsbank N.V.; SNS Bank N.V.; The Royal Bank of Scotland N.V. - Market share: 88 percent of total assets in the banking system. - Debt and baseline date: Latest data: December 2015. Source: supervisory data. Scope of consolidation: consolidated banking group. ### Channels of risk propagation / Methodology - Cash-flow-based using data on the time structure of undiscounted cash flows for up to one year. - Variants of LCR ratios by currency. - Test will be top-down. ### Risks and buffers - Risks: Funding liquidity; Market liquidity. - Buffers: Counterbalancing capacity; Central bank (standing) facilities (general framework only, excluding unconventional facilities). ### Tail shocks - Size of the shock: Bank run on deposits and dry up of wholesale funding markets, taking into account haircuts to liquid assets. - Run-off rates on funding sources calibrated to trigger a severe cumulative withdrawal of overall funding amount (about 20 percent over a three-month period). ### Regulatory standards - Liquidity gap, survival period. - Consistent with Basel III draft standards (LCR). ### Reporting format for results - Liquidity gap by bank. - Survival period in days by bank, number of banks that still can meet their obligations. *Italic: Source: cr1795 - 4. Risks and Buffers* --- ### Banking Sector: Contagion Risk ### Institutional perimeter - For domestic network analysis of contagion based on balance sheet data: six banks: ING Bank N.V.; Coöperatieve Rabobank U.A.; ABN AMRO Bank N.V.; N.V. Bank Nederlandse Gemeenten; Nederlandse Waterschapsbank N.V.; SNS Bank N.V. - For market based analysis: ING Bank N.V.; other G-SIBs; and other foreign banks. - Market share: 87 percent of total assets in the banking system. - Data and baseline date: December 2015 (balance sheet approaches); historical data for publicly traded banks up to June 2016 (market based approaches). ### Channels of risk propagation / Methodology - Network analysis using supervisory balance sheet data (Espinosa-Sole approach). - Network analysis using BIS data (Espinosa-Sole approach). - Analysis based on market data (Diebold-Yilmaz approach). ### Tail shock - Pure contagion: default of institutions, withdrawal of funding of failing institutions leading to partial replacement and fire sale of assets. ### Reporting format for results - Capital shortfall, by bank. - Capital shortfall, system wide. - Failure of individual institutions. - Interconnectedness measures. *Italic: Source: cr1795 - 4. Risks and Buffers* --- ### Appendix II. Risk Assessment Matrix — Selected Risks, Likelihood, and Impact - Economic fallout from political fragmentation and tighter or more volatile global financial conditions - Relative Likelihood: High - Impact: Medium - Key channels/effects: - Slower than anticipated external demand would negatively affect net exports, business confidence, and investment in the Netherlands; lower domestic growth would exacerbate credit risks. - A drop in stock and bond prices would affect insurers’ solvency and reduce coverage ratios in the pension sector; 28 percent of banks’ assets are in securities. - Renewed stress in global wholesale funding markets would increase funding costs for Dutch banks that rely on wholesale funding. - Depreciation of the euro could have negative impact on solvency in un-hedged institutions; could improve the financial strength of nonfinancial corporates (NFCs). - Weaker-than-expected global growth - Relative Likelihood: High/Medium (Euro Area and EMs); Low/Medium (China short-term/medium term) - Impact: Medium - Key channels/effects: - Close to 70 percent of exports are to European countries; Netherlands’ AEX index is the third most EM-exposed stock market in Europe. - Deflation could exacerbate real debt; continued low interest rates and low returns in equity markets could threaten solvency of insurance companies and prolong pension sector stress. - Lower-than-expected domestic growth - Relative Likelihood: Low - Impact: (described qualitatively) - Key channel/effect: - A reversal of the housing market recovery could weaken household balance sheets and dampen domestic demand; Dutch banks are highly exposed to households and NFCs. *Italic: Source: cr1795 - 4. Risks and Buffers* --- ### Appendix III. Methodology for the Construction of Macroeconomic Scenarios (housing market mapping) ### Purpose and approach - Country-specific layer includes shocks to domestic residential real estate markets. - Global macroeconomic model does not include country-specific housing markets; mapping is done by: 1. Analyzing how real estate market shocks affect consumption and investment. 2. Using the model to simulate effects of (exogenous) consumption and investment shocks as proxies for housing shocks. ### Residential real estate prices and consumption - Evidence: average marginal propensity to consume out of housing wealth is about 0.03–0.05 (widest plausible range 0.01–0.08). - Interpretation: consumption falls by €3–5 for every €100 decline in housing wealth. - Input data: - Nominal GDP in 2015 = €665 billion. - Estimated value of residential housing in the Netherlands = 3 X GDP = €2 trillion. - Nominal private consumption in 2015 = €302 billion. ### Example mapping used in stress scenario - Calculation reported: - A 15 percent decline in residential housing prices = €300 billion. - Marginal propensity to consume (MPC) = 0.2; impact on consumption = 0.02 X €300 billion = €6 billion. - Percentage decline in consumption = (6/302) X 100 = 2.0 percent — this is the private consumption shock corresponding to the country-specific layer in Table 3. - Note: mapping can be performed in nominal or real terms if housing prices and consumption are expressed in similar terms. ### Residential real estate prices and residential investment - Decline in housing prices from 2008-Q3 (peak) to 2013-Q2 (bottom) = 21 percent (in five years) — about 15 percent of which occurred from 2010-Q2 to 2013-Q2. - Investment composition assumption: - Business investment ≈ 70 percent of private investment. - Residential investment ≈ 30 percent of private investment. - A 31 percent decline in residential investment and a 13.4 percent decline in non-residential investment (over two years, as in 2009–10) would result in a 19 percent decline in private investment. *Italic: Source: cr1795 - 4. Risks and Buffers* --- ### Appendix IV. Satellite Models and Methodologies for Credit Loss Projections ### PD model specification - PD model specified as a panel auto-regressive distributed lag (p,p) structure: - LPD denotes the logistic transformation of the default probability (PD): ln(PD/(1-PD)) = LPD. - LPD dynamics include country-specific fixed effects (ηi, μi) and a vector of country-specific macroeconomic variables X = (RGDPG, RGDPG2, RLR), where: - RGDPG denotes the real GDP growth (quarterly) rate. - RGDPG2 is the square of the real GDP growth rate. - RLR is the real lending (annualized) interest rate prevailing in the quarterly period. - The dynamic panel model was estimated using quarterly data for the period 2004:Q1 through 2015:Q4. ### Estimated coefficients (presented in the source) - Explanatory Variable / Coeff - Change in PD (logit transformation, one-quarter lag): 0.19 - Real GDP growth (contemporaneous): -7.09 - Real interest rate (contemporaneous): 2.00 - Change in Real GDP growth (one-quarter lag): -9.15 - Change in Real GDP growth (two-quarter lag): -6.33 - Change in Real GDP growth squared (one-quarter lag): -224.90 - Change in Real GDP growth squared (two-quarter lag): -319.68 - Change in real interest rate (one-quarter lag): 0.07 - Change in real interest rate (two-quarter lag): -0.88 - Constant: -0.01 ### Projection rules and adjustments - Bank specific PDs for corporate loan portfolios were projected using the model’s predictions under the adverse scenario. - Projections account for country-specific fixed-effects and differences in end-2015 PDs, producing differentiated time changes in PDs across countries. - For exposures other than mortgages (governments, central banks, public sector entities, financial institutions, retail), PDs were assumed to increase by the same multiple as corporate loans. - Mortgage loan PD multiple = 0.85 times the corresponding corporate PD multiple. ### Adjustment of LGDs for mortgage loans in response to a decline in housing prices - Formula presented: - (1 - LGD2) = (H2 / H1) * (1 - LGD1) - Numerical example (Example 1 in source): - Case 1 baseline: House price H1 = 100 → H2 = 90 (adverse) - Loan L1 = 70 → L2 = 70 - LGD1 = 0.3 → LGD2 = ? - Recovery rate RR1 = 0.7; Recovery value R1 = 49 - Assume recovery value over house price is the same in both cases: 0.49. - Then R2 = 0.49 * 90 = 44.1; RR2 = 44.1 / 70 = 0.63; LGD2 = 0.37. - Example application in source: - If house prices decline 15 percent and LGD1 = 0.15; then LGD2 = 0.28. *Italic: Source: cr1795 - 4. Risks and Buffers* ### Appendix V. Contributions to Changes in Capital—Adverse ### Appendix V. Contributions to Changes in Capital—Adverse Scenarios ### Decomposition of changes in the CET1 capital adequacy ratio (CAR) - CAR is defined as Capital / RWA. - Changes over time in the CAR are expressed as: - Capital / RWA = CAR - The document gives the decomposition (symbols preserved as in source): - In the last expression, the first term on the right indicates the (partial) contribution of changes in capital (numerator) to variations in the CAR ratio; similarly, the second term is the contribution of changes in RWA (denominator). The third term captures the (joint) contribution of changes in RWA and capital to changes in the CAR ratio. The latter term is (very) small in size and can be added to the contributions of RWA, capital, or both. - For the construction of the decomposition chart, the joint effect (third term) is added to the contribution of changes in capital (numerator), yielding: - Contribution of Capital to CAR(1) = (expression shown in source as) - 11 1 1 1 CapitalRWA RWARWA tt tt tt t        - Contribution of RWA to CAR = (expression shown in source as) - 11 1 RWA (1) RWARWA tt tt X     - Change in RWAs term: - 11 1 RWACAR RWA tt t     ### Decomposition of capital evolution (assuming no capital injections) - Evolution of capital over time: - Capital_{t+1} - Capital_t = Capital_{t+1} - Capital_t = Profit_{t+1} - Dividends_{t+1} (symbols and order preserved) - Profit composition (as given): - Profit_{t+1} = Net interest income_{t+1} + Net gains on securities_{t+1} + Other items_{t+1} + Credit losses_{t+1} - Taxes_{t+1} - Other items_{t+1} include net fee and commission income, dividend income, and other items such as administrative/operational expenses. - Pre-impairment profit relation: - Profit_{t+1} can be written as: - Profit_{t+1} = Pre-impairment profit from previous year_{t+1} + Δ Net interest income_{t+1} + Δ Net gains on securities_{t+1} + Δ Other items_{t+1} - Δ Credit losses_{t+1} - Taxes_{t+1} - Pre-impairment profit from previous year_{t+1} is defined as: - Pre-impairment profit from previous year_{t+1} = Net interest income_{t} + Net gains on securities_{t} + Other items_{t} - Full decomposition of change in capital (symbols preserved as in source): - ( ) 1111 1111 Capital Pre-impairment profit from previous year Net interest income Net gains on securities Credit losses Other items Taxes Dividend. tttt tttt     - Sources of change in capital (numerator) and their contributions to changes in CAR (labels preserved): - Pre-impairment profits from previous year (1 Pre-impairment profit from previous year_{t+1}) - Losses or gains due to change in net interest income (1 Net interest income_{t+1} Δ) - Losses or gain due to market risk (securities) (1 Net gains on securities_{t+1} Δ) - Credit losses (1 Credit losses_{t+1}) - Taxes and other items (111 Other items Taxes Dividend_{ttt+1} Δ − −) ### Through‑the‑cycle adjustments to PDs and LGDs under the adverse scenario - Applied adjustments to all PDs/LGDs (labels and multipliers preserved): - REGPIT ΔPD = REGPIT · PD Δ = 0.5 (corporate and sovereign) - REGPIT ΔPD = REGPIT · PD Δ = 0.8 (retail) - REG ΔLGD = constant (all exposures) (Note: source presents these adjustments with the specific symbolic equalities: - 0.5 (corporate and sovereign) - 0.8 (retail) - constant (all exposures) and the symbolic relations REGPIT, REG, PD, LGD as in the source.) ### Liquidity stress tests: run‑off rates for liquidity outflows (In percent) - The document provides run-off rates by maturity buckets and instrument categories. Selected exact run‑off rates from the source table (labels and numeric values preserved): - Liabilities resulting from securities issued - Unsecured bonds due: 100 (for Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 75 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 50 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Unsecured bonds due (open maturity): 100 (Open maturity / Overnight), then "-" for subsequent buckets as shown in source - Hybrid bonds due: 100 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 75 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 50 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Hybrid bonds due (open maturity): 100 (Open maturity / Overnight), then "-" for subsequent buckets - Bonds eligible for the treatment set out in Article 129(4) or (5) of CRR due: 25 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 20 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 15 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Bonds as defined in Article 52(4) of Directive 2009/65/EC other than those reported to in item 1.1.3: 25 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 20 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 15 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Securitisations due: 35 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 25 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 20 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Short-term paper due: 100 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 75 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 50 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Of which: debt securities issued for retail only: 30 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 25 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 20 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Liabilities from secured lending and capital market driven transactions as defined in Article 192 of CRR, collateralised by Central Bank eligible assets: - Securities with a 0% risk weight: sequence shown in source as 55555555552220000 (digits presented exactly as in source) - Securities with a 20% risk weight: 20 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 15 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 10 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Bonds eligible for the treatment set out in Article 129(4) or (5) of CRR: 50 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 30 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 20 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Bonds as defined in Article 52(4) of Directive 2009/65/EC other than those reported to in item 1.2.1.3: 50 (Open maturity / Overnight through Greater than 3 weeks up to 4 weeks), 30 (Greater than 4 weeks up to 5 weeks through Greater than 2 months up to 3 months), 20 (Greater than 3 months up to 6 months through Greater than 12 months up to 2 years) - Table column and code headings are provided in the source (e.g., Code, ID, Item, and numeric column headers 010–180) and are preserved in the original. *Source: Appendix V. Contributions to Changes in Capital—Adverse Scenarios (document content provided).* ### Appendix VI. Liquidity Stress Tests ### Appendix VI. Liquidity Stress Tests ### Liquidity Stress Tests: Run-off Rates for Liquidity Outflows - Nonfinancial corporate bonds (ID 1.2.1.5): 50 (across the listed maturity buckets up to 2 years). - Residential mortgage backed securities of credit quality step 1 (ID 1.2.1.6): 50 (across the listed maturity buckets up to 2 years). - Other assets (ID 1.2.1.7): 75 for the earlier maturity buckets, and 50/30/30/30/30 for longer buckets as shown in the table. - Non-central bank eligible but tradable assets (ID 1.2.2): 100 for earlier maturity buckets, then 75/75/75/50/50/50/50 for longer buckets. - Equities listed on a recognised exchange, not self issued or issued by financial institutions (ID 1.2.2.1): same pattern as 1.2.2. - Gold (ID 1.2.2.2): same pattern as 1.2.2. - Other assets (ID 1.2.2.3): 100 (open maturity) and hyphen entries for several buckets as in the table. - Liabilities from deposits by customers that are not financial customers (ID 1.3): - By retail customers — Stable: 1, 2, 3, 5, 7, 8, 10, 15, 20, 20, 20, 20, 15, 15, 15, 15 (across successive maturity buckets as presented). - By retail customers — Less stable: 3, 5, 7, 10, 13, 15, 20, 25, 30, 30, 30, 30, 15, 15, 15, 15 (across successive maturity buckets as presented). - By retail customers (open maturity) — Stable: 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 1.0, 1.0, 1.0, 0.5, 1.5, 0.5, 0.5, 0.5, 0.5, 0.5. - By retail customers (open maturity) — Less stable: 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 3.0, 3.0, 3.0, 2.0, 6.0, 2.0, 2.0, 1.0, 1.0, 1.0. - By nonfinancial corporate customers (ID 1.3.2): 40 (across the listed maturity buckets up to 2 years). - By nonfinancial corporate customers (open maturity) — Operational: 0.1, 0.2, 0.2, 0.2, 0.2, 0.2, 1.3, 1.3, 1.3, 0.5, 1.5, 0.5, 0.5, 0.5, 0.5, 0.5. - By nonfinancial corporate customers (open maturity) — Other: 0.2, 0.4, 0.4, 0.4, 0.4, 0.4, 6.0, 6.0, 6.0, 1.5, 4.5, 2.0, 2.0, 1.0, 1.0, 1.0. - By other entities (ID 1.3.4): 30 (across the listed maturity buckets up to 2 years). - By other entities (open maturity): 0.2, 0.4, 0.4, 0.4, 0.4, 0.4, 6.0, 6.0, 6.0, 1.5, 4.5, 2.0, 2.0, 1.0, 1.0, 1.0. - Liabilities from customers that are financial customers (ID 1.4): - By credit institutions (ID 1.4.1): 100 (across the listed maturity buckets up to 2 years). - By credit institutions (open maturity): 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 10.0, 10.0, 10.0, 5.0, 10.0, 5.0, 2.0, 1.0, 1.0, 1.0. - By financial customers other than credit institutions (ID 1.4.2): 100 (across the listed maturity buckets up to 2 years). - By financial customers other than credit institutions (open maturity): 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 10.0, 10.0, 10.0, 5.0, 10.0, 5.0, 2.0, 1.0, 1.0, 1.0. - Of which: members of an institutional network (ID 1.4.3): 25 (across the listed maturity buckets up to 2 years). - Other cash-outflows (ID 1.7): 100 (earlier buckets) then 75/75/75/50/50/50/50 for longer buckets; open maturity pattern: 5.0, 5.0, 5.0, 5.0, 5.0, 10.0, 10.0, 10.0, 5.0, 10.0, 5.0, 2.0, 1.0, 1.0, 1.0. - Interest flows due (ID 1.8): 100 across all listed buckets (including open maturity indicated as 100). ### Liquidity Stress Tests: Roll-off Rates for Liquidity Inflows - Monies due from secured lending and capital market driven transactions (Article 192 of CRR) — central bank eligible assets and categories (IDs 2.1.1 through 2.1.1.7): entries appear in the table; categories include securities with a 0% risk weight, securities with a 20% risk weight, bonds eligible for Article 129(4) or (5), bonds as defined in Article 52(4) of Directive 2009/65/EC, nonfinancial corporate bonds, residential mortgage backed securities of credit quality step 1, and other assets. - Non-central bank eligible but tradable assets (ID 2.1.2) and subcategories (equities, gold, other assets) are listed in the table. - Monies due not reported in 2.1 from customers that are not financial customers (ID 2.2): - From retail customers (ID 2.2.1): 50 (across all listed maturity buckets up to 2 years). - From nonfinancial corporate customers (ID 2.2.2): 40 (across all listed maturity buckets up to 2 years). - From central banks (ID 2.2.3): 100 (across all listed maturity buckets up to 2 years). - From other entities (ID 2.2.4): 100 (across all listed maturity buckets up to 2 years). - Monies due not reported in 2.1 from financial customers (ID 2.3): 100 (across all listed maturity buckets up to 2 years). - From credit institutions (ID 2.3.1): 100 (across all listed maturity buckets up to 2 years). - From financial customers other than credit institutions (ID 2.3.2): 100 (across all listed maturity buckets up to 2 years). - Of which: members of an institutional network (ID 2.3.3): 100 (across all listed maturity buckets up to 2 years). - Other inflow items listed include FX-swaps maturing (ID 2.4), amounts receivable expected from contracts listed in Annex II of CRR (ID 2.5), paper in own portfolio maturing (ID 2.6), other cash inflows (ID 2.7), and interest flows received (ID 2.8) — these items appear in the inflows table with the roll-off rates by maturity bucket. ### Counterbalancing Capacity: Haircuts - Definitions: - A run-off rate is defined as the fraction of the liability amount maturing in a given period that is withdrawn by the claim holders. Sight deposits and contingent liabilities (committed credit and liquidity lines) are exceptions; run-off rates for those are defined with respect to the initial outstanding balance as noted. - A roll-off rate is defined as the fraction of the asset amount maturing in a given period that is converted into a cash inflow by the bank. Cash inflows from contingent credit/liquidity lines obtained by the bank are an exception and are defined with respect to the initial contracted amount. - Counterbalancing capacity haircuts (section 3) — selected items and haircuts: - Cash (03.1): 0 - Exposures to central banks (03.2): 0 - Unencumbered Central Bank eligible collateral (03.3): entries and subcategories listed; notable haircuts: - securities with a 0% risk weight (03.3.1) — subcategories representing claims on sovereigns, guaranteed by sovereigns, claims on or guaranteed by central banks, claims on or guaranteed by public sector entities, claims on or guaranteed by BIS/IMF/EU/multilateral development banks, and claims on or guaranteed by the EFSF/ESM: 3.8 (each of these sub-lines shows 3.8). - securities with a 20% risk weight (03.3.2) — sub-lines representing claims on sovereigns, guaranteed by sovereigns, claims on or guaranteed by central banks, claims on or guaranteed by public sector entities, and claims on or guaranteed by multilateral development banks: 25 (each of these sub-lines shows 25). - bonds eligible for the treatment set out in Article 129(4) or (5) of CRR (03.3.3): 39 (03.3.3 and 03.3.4 both list 39). - nonfinancial corporate bonds (03.3.5): 50 (03.3.5 and related sub-items indicate 50 and sub-lines for credit quality steps). - residential mortgage backed securities of credit quality step 1 (03.3.6): 30 - other central bank eligible assets (including credit claims) (03.3.7): 50 - Other unencumbered non central bank eligible, tradeable assets (03.4): 50 - equities listed on a recognised exchange, not self issued or issued by financial institutions (03.4.1): (listed under 03.4) - gold (03.4.2): (listed under 03.4) - Undrawn committed credit lines granted to the reporting institution (03.5): 0 (03.5 and sub-lines for members of the institutional network, intragroup entities, other entities are presented). - Net change of Counterbalancing Capacity (03.6) and Cumulated Counterbalancing Capacity (03.7) are shown in the table to compute available liquid resources under stress. *Sources: Authorities; and IMF staff.* ### References ### cr1795 - References ### Bibliographic entries - Altman, E., 2008, “Default Recovery Rates and LGD in Credit Risk Modeling and Practice: An Updated Review of the Literature and Empirical Evidence,” Chapter 7 in S. Jones and D. Hensher, editors, Advances in Credit Risk Modeling and Corporate Bankruptcy Prediction (New York: Cambridge University Press). - Beers, Bijlsma, and Mocking, 2015. “House Price Shocks and Household Savings: Evidence from Dutch Administrative Data,“ CPB Discussion Paper No. 299. - Bijlsma, Michiel, Nancy van Beers, Sander van Velhuizen, and Benedikt Vogt, 2015, “De Nederlandse Financieringstructuur in Perspectief”—Bijlage bij CPB Policy Brief 2015/14 “Een wereld zonder banken? Marktfinancierung en bankfinancierung in perspectief,”CPB Achtergronddocument, August. - Borio, Claudio, Leonardo Gambacorta, and Boris Hofmann, 2015, “The Influnce of Monetary Policy on Bank Profitability,“ BIS Working Paper No. 514, October. - Catalán, Mario, 2015, “Treatment of Liquidity Risks in Stress Tests,” IMF Guidance Note on Stress Testing,” Number 11, November (available upon request: mcatalan@imf.org). - Central Planbureau (CPB), 2014, “De Financiele Positie van het Mideen-en Kleinbedijf in Nederland,” CPB Notitie, June. - ______, 2014, “Financing SMEs” in CPB Financial Stability Report, pp. 24-31. - Espinosa-Vega, Marco, and Juan Sole, 2010, “Cross-Border Financial Surveillance: A Network Perspective,” IMF Working Paper No. 105, April. - Foglia, Antonella (2009), “Stress Testing Credit Risk: A Survey of Authorities’ Approaches,” International Journal of Central Banking, Vol. 5 No. 3, September. - Haldane, Andrew G., 2009, “Why Banks Failed the Stress Test,” speech delivered as Executive Director of Financial Stability at the Bank of England, Marcus-Evans Conference on Stress-Testing, London, February. - IMF, 2016, "Kingdom of the Netherlands—Netherlands—Selected Issues,” February, SM/16/4, Cor. 1 (Washington D.C.: International Monetary Fund). - ______, 2014, “SME Financing in the Netherlands,” Selected Issues Paper, Country Report 14/328. - Smets, Frank and Rafael Wouters, 2007. “Shocks and Frictions in US Business Cycles: A Bayesian DSGE Approach.” American Economic Review, vol. 97, no. 3, June. - van Beers, Nancy, Michiel Bijlsma, and Remco Mocking, 2015. “House Price Shocks and Household Savings: Evidence from Dutch Administrative Data,” CPB Discussion Paper No. 299. - Verbruggen, Johan, Remco van der Molen, Steven Jonk, Jan Kakes, and Willem Heeringa, 2015, “Effects of further reductions in the LTV limits,” Occasional Studies Vol. 13-2 (Amsterdam: De Nederlandsche Bank). - Vitek, Francis (2015), “Macrofinancial Analysis in the World Economy: A Panel Dynamic Stochastic General Equilibrium Approach,” International Monetary Fund Working Paper No. 15/227. *cr1795 - References* --- _Source: https://www.imf.org/-/media/files/publications/cr/2017/cr1795.pdf_