## 1polea2019005 ## Source details **Canonical URL:** [1polea2019005](https://www.imf.org/-/media/files/publications/cr/2019/1polea2019005.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2019/1polea2019005.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2019/1polea2019005.pdf.json) --- ### Executive summary — financial system structure and conditions - Banking sector dominance: - State-controlled and foreign-owned commercial banks account for about 60 percent of the financial sector (or 83 percent of the banking sector). - Financial system = 125 percent of the GDP (end-2016); banking sector ≈ three-quarters of system assets. - Ownership and systemic banks: - 12 OSIIs designated by PFSA as of December 2017; seven foreign-owned, six headquartered in the EU. (Since August 2018, OSIIs reduced from 12 to 11, with 6 foreign-owned OSIIs.) - Cooperative banks: - 553 cooperative banks; service about one quarter of the population; only 5 percent of financial system assets. - Focus on lending to agriculture (≈ half of cooperative lending) and SMEs. - Highly interconnected with affiliating commercial banks via an “Apex” network; two “Apex” commercial banks each account for < 1.5 percent of banking assets but are designated as OSIIs. - Financial soundness indicators: - Average Tier 1 ratio: 16.5 percent of RWA. - ROE: 10.4 percent (declining); EU average ROE = 6.1 percent. - FIAT: 0.44 percent on adjusted assets (introduced February 2016). - NPLs: 6 percent of total loans; overall coverage ratios ≈ 57 percent. - Liquid assets: over 20 percent of the total, mostly Polish government bonds. - Noted vulnerabilities: - Some affiliating banks and medium-sized domestic commercial banks show relatively high NPL ratios and low provisioning coverage. - Some cooperative banks show liquidity shortfalls on reported LCR; many exempted from LCR minimum due to participation in an Institutional Protection Scheme (IPS) under CRR provisions. ### FSAP risk analysis and stress-testing approach - Scope and methods: - Solvency stress testing (scenario-based and sensitivity analysis) and liquidity stress testing of commercial and cooperative banks. - Interbank, cross-sectoral, and cross-border contagion analysis using exposure and market data. - Consideration of second-round interbank contagion effects on banks’ solvency. - Coverage: commercial and cooperative banks accounting for 92 percent of banking sector assets included. - Scenarios: - Baseline: December 2017 WEO projections (cyclical upswing; inflation ≈ NBP target). - Baseline projections: Real GDP growth = 3.9 percent in 2018, moderating to 3 percent in 2020; inflation ≈ 2.5 percent through 2018–2020. - Adverse: extreme but plausible (external market stress, growth slowdown, domestic policy uncertainty) designed using IMF GFM model. - Adverse impacts: Output about 11.5 percentage points lower than baseline by 2020 (cumulative); long-term government bond yields rise by 280 basis points at peak relative to baseline. - Stress-test coverage: - Scenario-based macro stress tests covered 34 commercial banks (≈ 85 percent of banking sector assets). - Ownership of 34 banks: 52.7 percent foreign-owned; 38.3 percent state-owned. - Hurdle rates and dividend rules: - Dividend distribution disabled if total comprehensive income < 0 or if not in compliance with supervisory capital requirements or in restructuring. - Banks could distribute dividends only if leverage ratio > 5 percent and Tier 1 and total capital ratios were 150 basis points above regulatory requirements. ### Solvency stress test results and drivers - System-wide outcomes: - Aggregate CET1 declines from 16.2 percent to 12.9 percent of RWA in the adverse case. - Baseline system CET1 stabilizes around 15.2 percent. - Leverage ratio (group of 34 banks) falls from 10.2 percent to 8.3 percent in adverse case; 5 banks below 3 percent level. - 9 out of 34 banks could not meet regulatory minimum for one of the three capital ratios in the adverse scenario. - Asset share of undercapitalized banks: 20 percent of commercial banking system. - Necessary recapitalization need: 0.5 percent of GDP. - Main contributors to capital decline (quantified where provided): - Credit risk: accounts for 550 basis point decline in capital over 3 years. - Funding and interest rate risks: about 170 basis point decline over three years. - Valuation losses on debt portfolios: 90 basis point decline in first year; additional 0.2 percentage point decline when haircuts applied to HTM portfolios (robustness check). - Adverse sovereign spread: stressed sovereign bond spread rising by 280 basis points at its peak. ### Distributional and bank-type vulnerabilities - By bank size and type: - Medium-sized banks: CET1 declines from 13 percent to 7 percent; six out of fourteen medium-sized banks failed regulatory minimums; asset share of failing medium-sized banks = 18 percent of commercial bank assets. - OSIIs: CET1 declines from 17 percent to 14 percent; four OSIIs did not meet regulatory capital requirements and leverage ratio; asset share of under-capitalized OSIIs ≈ 12 percent of commercial bank assets. - Cooperative banks: exposed to credit and concentration risks; simultaneous default of five largest borrowers could breach minimum capital ratios for 61 percent of cooperative banking sector (5 percent of total banking system assets). ### Integrated solvency and interbank contagion findings - Interbank second-round impact: - Two simulation thresholds used (Tier 1 trigger = 6 percent and 3.5 percent). - Second-round impact reduced system Tier 1 capital ratio by 0.1 percentage points to 12.8 percent in both simulations. - For most commercial banks, interbank contagion impact was contained; one small bank with sizable interbank exposures experienced notable impact. - Network simulation setup: - Pair-wise on/off-balance sheet exposures between 34 commercial banks, 553 cooperative banks, and 352 foreign bank counterparties; simulations covered individual failures of 939 banks. - Assumptions: exposures treated as unsecured; first-round LGD = 100 percent; second-round funding loss = 35 percent of lost funding; asset sale haircut = 50 percent. ### Sensitivity analyses (single-factor shocks) - Exchange rate shock: - 30 percent depreciation of Zloty vs Swiss Franc applied (direct + indirect via mortgage satellite model) → Tier 1 capital declined by 120 basis points in combined shocks; all banks met Tier 1 capital ratio. - Interest rate shocks: - Upward shift 200 basis points; downward shift 100 basis points. - 200 bps funding shock + 500 bps sovereign bond yield rise (conservative across HFT/AFS/HTM) could lead to failure of one bank under conservative assumption; applying 500 bps only to HFT and AFS produced no bank failure. - Concentration shocks: - Simultaneous default of five largest borrowers per bank → Tier 1 ratio decline by 2.8 percentage points; failure of eight banks. ### Credit risk modelling and key drivers - Asset composition (commercial banks, 2017; BGK excluded): - Loans = 74 percent of total assets; Debt securities = 21 percent. - Loans by currency: Domestic 77.7 percent; Foreign 22.3 percent. - RWAs by risk factors: Credit risk 89.2 percent; Market risk 1.5 percent; Operational risk 7.8 percent; Others 1.5 percent. - Loan composition by sector: Mortgage 36.7 percent; Corporate 27.5 percent; Consumer 20.8 percent. - NPL projection methodology: - Satellite panel regressions for mortgage, consumer, and corporate NPL ratios; NPL projection inputs include GDP growth, unemployment, house prices, Swiss Franc/Zloty exchange rate, equity prices, wages, and interest rates. - Key estimation results (Annex IV excerpts): - Lagged dependent variable coefficients: Corporate 0.899***; Mortgage 0.883***; Consumer 0.821***. - Significant drivers with coefficients (selected): - Corporate: GDP growth (t-2) −0.0172*; Annual change in unemployment (t-1) 0.0280**; YoY change in equity prices −0.00151*. - Mortgage: Unemployment (t-2) 0.0329*; Swiss Franc/Zloty (t-2) 0.0863**; YoY house prices −0.00246**. - Consumer: Wage growth (t-2) −0.0100*; Long term interest rates 0.0273***. ### Liquidity stress testing — LCR and NSFR findings - Funding vulnerabilities (as of December 2017): - Loan-to-deposit ratio (commercial banks): about 102 percent (declining). - ≈ 70 percent of deposits have maturities < 30 days. - Only 31 percent of funding classified as stable (EBA reporting templates). - 67 percent of retail deposits classified as stable. - LCR exercise: - Four scenarios: baseline (CRR weights), adverse retail, adverse wholesale, combined (retail + wholesale + solvency haircuts). - Baseline system LCR: 149 percent; most banks between 100–200 percent; one medium-sized domestic bank failed marginally. - Retail event: system LCR > 100 percent; two medium-sized foreign banks failed; liquidity shortfall = 0.5 percent of 2017 GDP. - Wholesale event: system LCR = 120 percent; affected eight banks; liquidity shortfall = 0.5 percent of 2017 GDP. - Combined event: system LCR = 88 percent; affected 11 banks including three large ones; total shortfall < 3.6 percent of GDP (reported liquidity shortfall = 3.6 percent of 2017 GDP). - Bank counts tested: 34 (Large 5, Medium 14, Small 15). - Selected bank-count/asset-share outcomes by event (as presented): - Baseline (system LCR=149%): Banks with LCR <100: 1 (percent of commercial bank assets = 1.7). - Wholesale (system LCR=120%): Banks with LCR <100: 3 (percent of commercial bank assets = 9.5). - Combined (system LCR=88%): Banks with LCR <100: 36 (percent of commercial bank assets = 56.4). - LCR by currency: - Resilient in domestic currency in most scenarios (except extreme combined). - Not resilient in foreign currency: all banks failed each stress event in foreign currency; retail event had stronger adverse FX impact than wholesale. - Currency-by-currency LCR data did not fully reconcile with aggregate FX LCR data. - NSFR analysis: - Data quality concerns; mission’s NSFR results did not reconcile with PFSA survey data. - Mission reported System NSFR = 101.9. - Banks with NSFR <100: 38 (percent of commercial bank assets = 56.9). - Liquidity shortfall (NSFR) = 6.0 percent of 2017 GDP. - Authorities’ PFSA calculation reported NSFR ≈ 120 percent. ### Contagion, interconnectedness and systemic channels - Interbank network simulations (Espinosa-Vega and Solé approach): - When only commercial banks considered, no more than one additional failure induced; outward spillover index ≲ 4 percent of Tier 1 capital. - Funding shocks drove most contagion effects. - Combined commercial + cooperative simulations: - Average induced failures from medium-sized bank events rose to 38; maximum induced failures rose to 338. - Cooperatives’ failures driven by credit losses from affiliating bank defaults and structural features; outward spillover index increased marginally given small size of cooperatives. - Foreign counterparties: - Including 352 foreign counterparties, failures of only three foreign banks induced an additional failure. - Average total assets of additionally failed banks in those cases = 0.5 percent of total system assets. - Sovereign-bank nexus: - Banks’ sovereign holdings increased (commercial banks’ holdings rose from ~20 percent of RWA at Dec 2015 to ~25 percent at Dec [date truncated in source]). - General government claims = 20 percent of total assets of other MFIs by end-2017 (up from ≈17 percent pre-FIAT). - By end-2017 government directly/indirectly controlled 40 percent of assets of deposit-taking institutions (up from 26 percent in 2015). - A 500 bps increase in government yields reduces aggregate Tier 1 capital by about 2.2 percentage points under simple assumptions; adjusting for 12 percent relative increase of sovereign holdings post-FIAT reduces impact to 2.0 percentage points. - Market-based connectedness (Diebold & Yilmaz): - Large banks are net transmitters of shocks domestically. - Polish domestic connectedness relatively contained and low compared with Spain FSAP indexes. - Cross-border market-pricing results indicate Polish banking system is generally a net recipient of foreign shocks; exception: net transmitter to Portuguese banking system (consistent with Portuguese exposures to Poland). ### FX mortgage portfolio — background, stress, and policy implications - Size and evolution: - FX mortgages (mostly Swiss franc): from 2 percent of GDP (~60 percent of housing loans) end-2004 to 11 percent of GDP (~70 percent of housing loans) by end-2008. - Stock declined from early 2012; amortization rate ≈ 8 percent annually recently; FX mortgage stock = 6½ percent of GDP by end-2017. - Risk profile and regulatory responses: - FX mortgage NPLs remained below 4 percent historically; 2015 Swiss franc appreciation raised LTVs and debt service burdens but NPLs rose only moderately. - Prudential measures tightened since 2006; effective ban on new FX loans to unhedged borrowers from 2013. - 2017 FSC recommendations: increase FX mortgage risk weight to 150 percent; impose 3 percent systemic risk buffer; require additional capital for banks with FX mortgage portfolios (most in effect late 2017/early 2018). - Stress scenarios and cost estimates: - Under main adverse scenario, three-year cumulative cost ranges between 7.3 and 8.1 billion zloty (0.2–0.3 percent of RWAs annually). - Baseline/ongoing conversion assumptions: - Cumulative amount “at risk” after shocks ≈ 26 percent of end-2017 FX mortgage stock; conversion could take 4½ years. - Haircuts needed: 25–30 percent to equalize monthly installments pre- and post-conversion. - 30 percent zloty depreciation scenario: - One-year cost = 3 to 3.3 billion zloty (22–25 percent of net profits and other comprehensive income). - Total amount “at risk” after one-time shock ≈ 75 percent of end-2017 FX mortgage stock; conversion could take 11½ years. - Haircut needed: 29 percent to equalize monthly installments pre- and post-conversion. - FSAP assessment and policy guidance: - Risks associated with FX mortgage portfolio have declined; no systemic policy response warranted. - Solutions should be case-by-case and risk-based; any policy action for non-financial-stability objectives should minimize adverse impacts on bank profitability and financial stability. - Scenario-based “Restructuring Fund” analysis presented upper (“no conversion”) and lower (“fast conversion”) bounds with contributions capped at 0.5 percent of FX mortgage stock and first-year cap of 2¾ billion zloty in one proposal. ### Key policy recommendations (selected, verbatim where applicable) - Address identified weaknesses in medium-sized banks and OSIIs, including affiliating banks for cooperative banks; strengthen supervision and oversight of relevant OSIIs and other weak banks. - Agency: PFSA; Time: I/NT. - Replace the FIAT with a Financial Activities Tax (FAT) to reduce incentives for banks to hold sovereign debt and to address stability and efficiency concerns. - Agency: MoF; Time: I/NT. - Improve enforcement of common LCR parameters and monitoring of FX liquidity risk; ensure banks’ readiness for CRR II implementation timeframe. - Agency: PFSA; Time: NT. - Improve data for credit risk analysis: collect PD and LGD data and move towards PD/LGD-based models, especially with IFRS 9. - Agency: PFSA, NBP; Time: NT. - Improve collection of institution-by-institution counterparty-specific asset- and liability-side exposures and loss-absorbing capacity to enable better intra-sector and cross-sector contagion analyses. - Agency: PFSA, NBP; Time: I/NT. - Improve collection of Polish banks’ asset- and liability-side cross-border exposures and collect more granular data on secured and unsecured interbank exposures. - Agency: PFSA, NBP; Time: NT. - Time frame legend reproduced from source: - C = continuous; I (immediate) = within one year; NT (near term) = 1–3 years; MT (medium term) = 3–5 years. ### Selected preserved statistics and metrics (from source) - Financial system size: 125 percent of GDP (end-2016). - Public debt-to-GDP: about 50 percent (capped at 60 percent by constitution). - Banking sector coverage in stress tests: 92 percent of banking sector assets. - Commercial banks in scenario-based macro stress tests: 34 banks (~85 percent of banking sector assets). - Ownership shares (34 commercial banks): 52.7 percent foreign-owned; 38.3 percent state-owned. - Asset composition (commercial banks, 2017): Loans 74 percent; Debt securities 21 percent. - Loan currency split: Domestic 77.7 percent; Foreign 22.3 percent. - RWAs by risk factors: Credit risk 89.2 percent; Market risk 1.5 percent; Operational risk 7.8 percent; Others 1.5 percent. - Loan composition by sector: Mortgage ~36.7 percent; Corporate ~27.5 percent; Consumer ~20.8 percent. - Adverse scenario cumulative output loss: about 11.5 percentage points by 2020. - FIAT asset tax rate: monthly rate 0.0366 percent (or 0.44 percent per year). - Dividend distribution leverage threshold: leverage ratio > 5 percent; Tier 1 and total capital ratios required to be 150 basis points above regulatory requirements to distribute dividends. - System LCR (baseline): 149 percent; Combined extreme LCR = 88 percent; combined extreme liquidity shortfall < 3.6 percent of GDP. - System NSFR (mission result): 101.9; NSFR shortfall = 6.0 percent of 2017 GDP (per mission estimates). - FX mortgage stress estimate: three-year cumulative cost between 7.3 and 8.1 billion zloty (0.2–0.3 percent of RWAs annually); one-year cost under 30 percent depreciation = 3 to 3.3 billion zloty. *Source: IMF staff calculations and Poland FSAP (extract provided).* ### EXECUTIVE SUMMARY ___________________________________________________________________________ 6 ### EXECUTIVE SUMMARY ### Financial system structure and financial conditions - Poland’s financial system is dominated by the banking sector, with significant state participation and foreign ownership. - State-controlled and foreign-owned commercial banks account for about 60 percent of the financial sector (or 83 percent of the banking sector). - The financial system is 125 percent of the GDP (end-2016) with the banking sector representing about three-quarters of system assets. - Among the 12 domestic systemically important banks (OSIIs) designated by the PFSA as of December 2017, seven are foreign-owned, with six of these headquartered in the European Union. (Since August 2018, the number of OSIIs reduced from 12 to 11, with 6 foreign-owned OSIIs.) - Cooperative banks: - Service about one quarter of the population, with only five percent of financial system assets. - There are 553 cooperative banks, focusing on lending to agriculture (providing around half the total) and SMEs. - Cooperative banks are highly interconnected with their affiliating commercial banks through an “Apex” network structure; the two “Apex” commercial banks each account for less than 1.5 percent of banking assets but are designated as OSIIs. - Financial soundness indicators: - Average Tier 1 ratio: 16.5 percent of RWA. - ROE: 10.4 percent (declining), above EU average of 6.1 percent ROE. - FIAT: 0.44 percent on adjusted assets (introduced February 2016). - NPLs: 6 percent of total loans; overall coverage ratios about 57 percent. - Liquid assets: over 20 percent of the total, mostly Polish government bonds. - Some affiliating banks and medium-sized domestic commercial banks appear to have relatively high NPL ratios and low provisioning coverage. - Some cooperative banks show liquidity shortfalls based on reported LCR; most of those are exempted from LCR minimum by the competent authority due to participation in an Institutional Protection Scheme (IPS) under CRR provisions. ### FSAP risk analysis and stress testing approach - The FSAP conducted a comprehensive set of stress tests and contagion analyses, including: - Solvency stress testing (scenario-based and sensitivity analysis) and liquidity stress testing of commercial and cooperative banks. - Interbank, cross-sectoral, and cross-border contagion analysis using both exposure and market data. - Consideration of second-round impact of interbank contagion on banks’ solvency positions. - Scope included commercial and cooperative banks; methodology details and additional analyses are provided in annexes (e.g., PD/LGD collection recommendation, contagion methodology). ### Solvency stress test results and vulnerabilities - Aggregate system resilience: - System CET1 declines from 16.2 percent to 12.9 percent of RWA in the adverse case, driven by loan loss provisions, valuation losses on debt securities, and funding interest rate risks. - Distributional vulnerabilities: - Some medium-sized banks would come under pressure; necessary recapitalization need at around 0.5 percent of GDP. - The cooperative banking sector is exposed to credit and concentration risks; this is evidenced by simulations such as default of the five largest borrowers. - Recommendations arising from solvency findings: - Sustain supervisory attention towards weaknesses in medium-sized banks and OSIIs, including affiliating banks of cooperative networks, given their importance in the banking system. ### Liquidity stress test results - Overall assessment: - Banking system appears resilient to short-term liquidity risks in local currency. - Adverse retail event had a limited impact; the wholesale event affected some large banks. - Combined extreme event compounded costs that totaled less than 3.6 percent of GDP. - LCR and NSFR findings: - LCR by currency: resilient in local currency but not in foreign currency (FX). - NSFR tests indicate vulnerability to more protracted events, though there is uncertainty about the robustness of data provided to the FSAP team. - Recommendations from liquidity findings: - Improve monitoring of FX liquidity risks. - Ensure banks’ readiness for the upcoming NSFR implementation timeframe. ### Contagion and systemic risk analysis - Domestic and cross-border contagion risks are generally limited with some material exceptions. - Interbank connectedness: - Any destabilization to affiliating banks could cause severe distress to the affiliated cooperative banks; reinforces need to address weaknesses in affiliating banks. - Sovereign-bank nexus: - Banks’ large holdings of sovereign bonds and substantial presence of state-controlled banks produce a significant sovereign-bank nexus. - Commercial banks’ holdings of government securities increased from around 20 percent of RWA at December 2015 to around 25 percent at December (date truncated in source). - Policy implication: - To avoid incentives for banks to hold sovereign debt, the FSAP team recommends redesigning the Financial Institution Asset Tax (FIAT). ### Key recommendations (Table 1 summary) - 1. Address identified weakness in medium-sized banks and OSIIs, including the affiliating banks for cooperative banks, given their systemic importance in the banking system. - Agency: PFSA - Time: I/NT - 2. Replace the FIAT with a Financial Activities Tax (FAT), addressing both stability and efficiency concerns. - Agency: MoF - Time: I/NT - Note: Recommendation consistent with Poland Article IV Staff Reports (2016 and 2017). - 3. Improve the enforcement of common parameters of the LCR and the monitoring of FX liquidity risk and ensure banks’ readiness for the upcoming CRR II implementation timeframe. - Agency: PFSA - Time: NT - 4. Improve data for credit risk analysis by collecting PD, LGD data and move towards PD, LGD-based models. - Agency: PFSA, NBP - Time: NT - 5. Improve the collection of institution-by-institution counterparty-specific data for banks and non-banks on asset and liability-side exposures and loss-absorbing capacity to allow for improved analyses of intra-sector and cross-sectoral contagion. - Agency: PFSA, NBP - Time: I/NT - 6. Improve the collection of Polish banks’ asset- and liability-side cross-border exposures to foreign banks. Collect more granular data on banks’ secured and unsecured interbank exposures. - Agency: PFSA, NBP - Time: NT - Time Frame legend: - C = continuous - I (immediate) = within one year - NT (near term) = 1–3 years - MT (medium term) = 3–5 years *Source: IMF — Poland FSAP Executive Summary (selected content).* ### 2017. At about 50 percent of GDP, and capped at 60 percent by the constitution, the public debt to ### 1polea2019005 - 2017. At about 50 percent of GDP, and capped at 60 percent by the constitution, the public debt to ### Overview - At about 50 percent of GDP, and capped at 60 percent by the constitution, the public debt-to-GDP ratio is moderate, implying limited sovereign risks in Poland at present. - Sovereign exposures can mutually reinforce fiscal and financial vulnerabilities, and where the securities are of long duration, interest rate rises may further amplify the shocks. ### Risk Analysis in the Poland FSAP - The FSAP risk and vulnerability analysis provided an in-depth assessment of financial system resilience to adverse shocks and contagion. - Components of the risk analysis included: - Top-Down (TD) solvency and liquidity stress testing of the banking sector. - Domestic and cross-border contagion analysis. - Single-factor sensitivity analyses. - Coverage: commercial and cooperative banks accounting for 92 percent of banking sector assets were included in the analysis. - Stress test risk types covered: credit, market (equity, exchange rate, and interest rate risks), sovereign (repricing and spread risks), and funding interest rate risks. - Liquidity metrics assessed: LCR and estimated NSFR, both at aggregate level and by currency. - Contagion analysis examined interbank, cross-sectoral, and cross-border interconnectedness and bank-sovereign nexus. - Integration: failed banks from solvency stress tests were used as initial triggers in interbank network analysis to capture second-round impacts on bank capital. - Caveats highlighted: - FSAP stress tests are macroprudential diagnostic tools and not macroeconomic forecasts. - Scenarios capture tail events based on historical distributions; actual crises may deviate from historical relationships. - Results depend on availability and quality of data provided. ### Banking Sector Solvency Stress Tests — Scope and Coverage - Solvency stress tests and sensitivity analyses covered commercial and cooperative banks, accounting for 92 percent of banking sector assets. - Scenario-based macro stress tests covered 34 commercial banks, accounting for about 85 percent of total banking sector assets. - Ownership composition of the 34 commercial banks: about 52.7 percent foreign-owned, 38.3 percent state-owned, remainder privately-owned (domestic). - Cooperative banks: not systemic by assets but included in sensitivity analyses due to concentrated local exposures. - Exemptions: state-owned development bank BGK is explicitly exempted from CRD/CRR requirements and excluded from the stress testing sample. ### Macroeconomic Risks and Scenarios - Two scenarios used: - Baseline scenario: based on December 2017 WEO projections; reflects strong cyclical upswing in GDP with inflation broadly at NBP’s target. - Adverse scenario: extreme but plausible, reflecting external financial market stress, growth slowdown, and domestic policy uncertainty; designed using IMF’s GFM model. - Baseline projections (selected): - Real GDP growth projected: 3.9 percent in 2018, moderating to 3 percent in 2020. - Inflation around target of 2.5 percent throughout 2018–2020. - Adverse scenario narrative: - Financial market stress and tightening in systemic economies; heightened uncertainty and confidence losses in Europe and the United States; protectionist measures; constrained macro policy responses; domestic policy slippages and weakened institutions. - Adverse scenario impacts: - Output about 11.5 percentage points lower than the baseline by 2020 (cumulative GDP level loss relative to baseline). - Long-term government bond yields rise by 280 basis points compared to the baseline at their peak due to a potential rating downgrade. - Table 2. Poland: Macroeconomic Scenarios for Stress Tests (In percent unless otherwise specified) - Adverse Scenario / Baseline Scenario — 2018 / 2019 / 2020 - Real GDP (2017=100): 98.6 / 97.0 / 99.0 — 103.9 / 107.3 / 110.5 - Real GDP growth: -1.4 / -1.6 / 2.0 — 3.9 / 3.3 / 3.0 - Inflation (CPI): 2.0 / 0.0 / - — 2.6 / 2.5 / 2.5 - Monetary Policy rate: 1.0 / 0.8 / 0.1 — 1.9 / 2.9 / 3.9 - Short-term money market rate: 2.0 / 2.0 / 1.3 — 1.9 / 2.9 / 3.9 - Long term government bond yield: 6.2 / 7.0 / 7.2 — 3.8 / 4.2 / 4.5 - Unemployment rate: 5.7 / 6.7 / 6.5 — 4.0 / 3.9 / 3.8 - Exchange rate (Zloty/SF): 4.2 / 4.3 / 4.3 — 3.8 / 3.8 / 3.8 - Nominal GDP growth: 0.6 / -1.6 / 1.1 — 5.9 / 5.8 / 5.3 - Note: adverse scenario severity benchmarked using EU crisis episodes; baseline is December 2017 WEO. ### Approach for Scenario-Based Solvency Stress Test - Framework: IMF’s internally developed solvency stress testing framework capturing credit risk, sovereign risk (repricing and spread), market risk (interest rate, FX, equity), funding and interest rate risk on the banking book. - Balance sheet and RWAs: - Balance sheet growth assumed quasi-static allocation; composition of funding unchanged. - Balance sheet size projected to grow in line with nominal GDP paths, with cap on deleveraging in adverse scenario. - RWAs adjusted for balance sheet growth, additional risk weights from newly defaulted loans, exchange rate movements, provisions, and differentiated risk weights for domestic and foreign currency mortgage portfolios. - Income projections: - Total comprehensive income reflected net interest income (base and interest rate risk components), non-interest income and expense, trading income and losses, other comprehensive income, loss provisions, income tax and the FIAT. - FIAT description: starting in February 2016, levied on the total value of assets (less PLN 4 billion, own funds, and purchased sovereign debt) at a monthly rate of 0.0366 percent (or 0.44 percent per year). Exclusions: state-owned banks (relevant for BGK with legal status of state enterprise) and banks under recovery proceedings, in receivership, in liquidation, or those who filed for bankruptcy. Affiliating banks pay practically no asset tax due to further exclusions of their exposure to cooperative banks. - Liability interest rate (funding rate) shocks estimated via satellite models; NPL ratios projected using satellite models by sector (household mortgages, consumer credit, corporate credit). - Dividend distribution assumptions: - Banks not allowed to distribute dividends if total comprehensive income was below zero, or if not in compliance with supervisory capital requirements, or in restructuring. - Banks could only distribute dividends if leverage ratios were above 5 percent, and Tier 1 and total capital ratios were 150 basis points above regulatory requirements. - A tiered dividend distribution policy accounted for banks’ FX exposures, following PFSA regulations. - Hurdle rates: - Results benchmarked against two sets of hurdle rates: 1) capital ratio standards including Pillar I and Pillar II requirements (similar to NBP Financial Stability Review). 2) Pillar I and Pillar II requirements and OSII buffer (similar to other European FSAPs). - Leverage ratio considered as additional metric. - Table 3. Poland: Hurdle Rates for Solvency Stress Tests - Hurdle rate excluding OSII Buffers / Hurdle rate including OSII Buffers - Total capital ratio: 8 / 0 to 5.81 — 8 / 0 to 5.81 / 0 to 0.75 - Tier 1 capital ratio: 6 / 0 to 4.36 — 6 / 0 to 4.36 / 0 to 0.75 - CET1 capital ratio: 4.5 / 0 to 3.25 — 4.5 / 0 to 3.25 / 0 to 0.75 - Note: Pillar II buffers are designed to mitigate risks in the FX mortgage portfolios. ### Credit Risk Analysis and Estimation - Credit risk is the most important risk factor for the banking system. - Asset composition (commercial banks; BGK excluded; Date: 2017): - Loans represented about 74 percent of total assets of commercial banks. - Debt securities: 21 percent. - Loans by currency: Domestic 77.7 percent; Foreign 22.3 percent. - RWAs by risk factors: Credit risk 89.2 percent; Market risk 1.5 percent; Operational risk 7.8 percent; Others 1.5 percent (as presented in figure breakdown). - Loan composition by sector: Corporate 27.5 percent; Mortgage 36.7 percent; Consumer 20.8 percent; Central Bank 3.4 percent; Government 3.2 percent; Credit institutions 2.6 percent; Other financial 2.6 percent; Other 1.5 percent (as presented in figure breakdown). - Mortgage loans accounted for about 37 percent of total loan portfolios; corporate and consumer loans follow. - By currency, more than three-quarters of loans were in domestic currency (78 percent); 22 percent foreign currency, majority concentrated in mortgage segment dominated in Swiss Francs. - Credit risk accounted for 89 percent of RWAs, followed by market risk (1 percent) and operational risk (8 percent) in the narrative. - Modelling approach: - Satellite panel regression models used to project NPL ratios for household mortgages, consumer credit, and corporate credit. - Provisioning ratios computed based on historical provisions by loan type. - Majority of banks operate under Basel II standardized approach: thirty-one out of 34 banks applied exclusively the standardized approach; 3 banks applied IRB partially. - Macrofinancial drivers of credit risk (per satellite models): - Corporate default rates: sensitive to declines in real GDP growth and equity prices and rises in unemployment. - Mortgage default rates: sensitive to unemployment rate, house price changes, and Swiss Franc/Zloty exchange rate movements (exchange rate risks associated with Swiss Franc appreciation captured in mortgage satellite model). - Consumer loan default rates: associated with wage growth and interest rates. ### Key Statistics and Findings (preserved from source) - Public debt-to-GDP: about 50 percent (capped at 60 percent by constitution). - Banking sector coverage in stress tests: 92 percent of banking sector assets. - Commercial banks in scenario-based macro stress tests: 34 banks (~85 percent of banking sector assets). - Ownership shares (34 commercial banks): 52.7 percent foreign-owned; 38.3 percent state-owned. - Asset composition (commercial banks, 2017): Loans 74 percent; Debt securities 21 percent; Non interest-earning assets and other assets remainder. - Loan currency split: Domestic 77.7 percent; Foreign 22.3 percent. - RWAs by risk factors: Credit risk 89.2 percent; Market risk 1.5 percent; Operational risk 7.8 percent; Others 1.5 percent. - Loan composition by sector (figures cited): Mortgage ~36.7 percent; Corporate ~27.5 percent; Consumer ~20.8 percent. - Adverse scenario output loss: about 11.5 percentage points lower than baseline by 2020. - Long-term government bond yield peak increase in adverse scenario: 280 basis points relative to baseline. - FIAT asset tax rate: monthly rate of 0.0366 percent (or 0.44 percent per year). - Dividend distribution leverage threshold: banks must have leverage ratios above 5 percent to distribute dividends. - Dividend distribution capital buffer threshold: Tier 1 and total capital ratios were required to be 150 basis points above regulatory requirements for dividend distribution. *Source: IMF staff calculations and Poland FSAP (extract provided).* ### 21. Market risk analysis captured the valuation risks of bonds due to changes in risk-free ### 21. Market risk analysis captured the valuation risks of bonds due to changes in risk-free interest rates and credit spreads, foreign exchange risks and equity risks from net open positions ### Market risk framework and scope - Analysis covered: - Interest rate risks and spread risks on sovereign and corporate debt securities in the held-for-trading (HFT) book and the available for sale (AFS) book. - Foreign exchange risks and equity risks from net open positions. - Regulatory capital impact: - Losses from HFT portfolios impacted regulatory capital through net profits. - Losses for AFS portfolios affected capital through other comprehensive income. - Foreign exchange rate risks and equity risks from net open positions influenced net profits directly. ### Debt securities composition and portfolio metrics - Debt securities mainly held in the available for sale (AFS) book and dominated by sovereign exposures. - Portfolio composition (commercial banks, BGK excluded): - Sovereign debt accounted for 87 percent of the debt securities portfolio. - Central Bank bills accounted for 5 percent of the debt portfolios. - Debt securities by accounting treatment: AFS 80 percent, hold-to-maturity (HTM) 11 percent, held-for-trading (HFT) 9 percent. - Average duration of the debt portfolio: 1.2 years (with sizable variations across banks). - NBP estimate: share of domestic bond holdings = 94 percent of total bond holdings. - Note: Information based on commercial banks; state-owned development bank BGK is excluded. Date: 2017. ### Valuation methodology for interest rate and spread risks - Market valuation losses from interest rate and spread risks derived using a modified duration approach. - Two valuation channels for HFT and AFS books: 1. Re-pricing losses due to shocks to the risk-free interest rate (measured by the policy rate in Poland). 2. Valuation impact due to shocks to the spread of debt securities. - Spread construction: - Sovereign spread = 10-year Polish government bond yield minus 10-year German government bond yield. - Corporate bond yield = average bond yield for nonfinancial corporate debt available from Bloomberg. - Robustness check: conservative assumption applying haircuts to HTM securities. ### Foreign exchange and equity risk treatment - Direct impact analyzed through banks’ net open positions in foreign currencies and equities. - FX net open positions captured all foreign currency positions in the banking and trading book, mainly in Swiss Francs, the US dollar, and the Euro. - Equity open position reflected banks’ equity holdings. - Shocks to exchange rates and equity prices were applied to these open positions to evaluate gains or losses. - Exchange rate risks were captured through two channels in scenario-based solvency stress testing: 1. Direct exchange rate impact on banks’ net open positions in FX. 2. Indirect exchange impact (Swiss Franc appreciation) on credit risk for mortgage portfolios through satellite models. ### Interest rate and funding risks in the banking book - Assessment method: maturity gap analysis accounting for liability (funding) interest rate shocks. - Net interest income impact estimated by measuring gap between interest sensitive assets and liabilities across repricing buckets from less than one month to end of three-year stress horizon. - Liability interest rates estimated using satellite models with dynamic panel regression techniques. - Effective interest rates on liabilities sensitive to 3-month WIBOR and 10-year government bond yields. - Under the adverse scenario: - Relationship between changes in funding and lending rates estimated using quantile regressions. - Imperfect pass-through implies tightening in interest rate margins for banks. - Detailed estimations on funding rates in Annex V. ### System-wide scenario-based solvency stress test results - Baseline scenario: - Aggregate CET1 ratio stabilizes at around 15.2 percent. - Leverage ratio (Tier 1 to total assets) remains at around 10 percent for the group of 34 commercial banks. - Slight decline in capital ratios partly due to rise in risk weighted assets and valuation losses in the AFS portfolio from a rise in interest rate. - Adverse scenario: - CET1 ratio declines from 16.2 percent to 12.9 percent for the system. - 9 out of 34 banks could not meet the regulatory minimum for one of the three capital ratios (total capital, Tier 1 capital or CET1 capital ratios). - Leverage ratio declines from 10.2 percent to 8.3 percent for the group of 34 commercial banks, with 5 banks below the 3 percent level. - Asset share of undercapitalized banks: 20 percent of the commercial banking system. - Necessary recapitalization: 0.5 percent of GDP. ### Main contributors to capital decline under the adverse scenario - Overall drivers (with quantified contributions where provided): - Credit risk: - Dominant driver reflecting loan book (three-quarters of assets). - Accounts for 550 basis point decline in capital over the 3-year stress testing horizon. - Corporate loan losses driven by decline in GDP growth and equity prices, and rise in unemployment. - Mortgage loan losses driven by rise in unemployment, decline in house prices, and depreciation of Zloty/Swiss Franc exchange rate. - Consumer loan losses driven by decline in wage growth and rise in interest rates. - Funding and interest rate risks: - Contributed about 170 basis point decline in capital over three years. - Funding shocks from rise in 3-month WIBOR and long-term government bond yield led to decline in net interest rate margins. - Margins further compressed by formation of new NPLs that did not generate interest income in the adverse scenario. - Valuation losses on debt portfolios: - Led to a decline in capital by 90 basis point in the first year (through other comprehensive income). - Adverse scenario stressed sovereign bond spread rising by 280 basis points at its peak. - Valuation losses applied to both HFT and AFS portfolios using modified duration approach. - Robustness check: applying haircuts to HTM portfolios (11 percent of debt security holdings) produced an additional decline of capital ratios by 0.2 percentage points. ### Solvency results by bank type - By bank size: - Medium-sized banks: CET1 ratio declined from 13 percent to 7 percent. - Six out of fourteen medium-sized banks failed to meet regulatory minimum capital requirement. - Asset share of failing medium-sized banks: 18 percent of commercial bank assets. - OSIIs (other systemically important institutions): - CET1 ratio declined from 17 percent to 14 percent. - Four OSIIs did not meet regulatory capital requirements and the leverage ratio. - Asset share of under-capitalized OSIIs: about 12 percent of commercial banks assets. ### Integrated solvency and interbank contagion analysis - Integration conducted with NBP using confidential interbank counterparty information. - Two simulation thresholds for triggering and secondary failures: 1. Trigger/secondary threshold = 6 percent (Tier 1 capital ratio). 2. Trigger/secondary threshold = 3.5 percent. - Second-round impact from interbank contagion: - Reduced system-wide Tier 1 capital ratio by 0.1 percentage points to 12.8 percent in both simulations. - For most commercial banks, impact was contained; exception: one small bank with sizable interbank exposures experienced notable impact. ### Sensitivity analysis — Commercial banks - Single-factor shocks assessed: exchange rate, interest rate, and concentration (counterparty) risks. - Exchange rate shock: - 30 percent depreciation of the Zloty against the Swiss Franc applied to net open positions (direct) and mortgage loan portfolios via satellite model (indirect). - Tier 1 capital ratio declined by 120 basis points in the combined shocks, with all banks meeting their Tier 1 capital ratio. - Interest rate shocks: - Upward shift: 200 basis points. - Downward shift: 100 basis points. - A 200 bps increase in funding shock and a 500 bps rise in sovereign bond yield (conservative assumption applied to all HFT, AFS, and HTM portfolios) could lead to the failure of one bank. - Two sovereign bond yield simulations: 1. 500 bps rise applied to HFT and AFS portfolios according to valuation standards (no bank failure). 2. 500 bps rise applied to HFT, AFS, and HTM portfolios (conservative); HTM not required to be marked-to-market per accounting standards. - Concentration risk shocks: - Simultaneous default of the five largest borrowers for each bank would lead to: - Decline in Tier 1 ratio by 2.8 percentage point. - Failure of eight banks. ### Sensitivity analysis — Cooperative banks - Conducted by NBP with single-factor shocks aligned with FSAP team. - Shocks included counterparty/concentration, credit, and interest rate shocks: - Counterparty/concentration: simultaneous default of five largest non-financial borrowers for each bank and for sector. - Credit shocks: 1. Increase in impaired loan coverage of cooperative banks to same level as commercial banks. 2. 50 percent impairment on a given percentage of performing loans to non-financial sector. - Interest rate shocks: upward shift by 200 basis points, downward shift by 100 basis points, and 300 basis points parallel widening of government bond yield. - Findings: - Cooperative sector generally resilient to interest rate risks (limited impact on capital ratios). - Credit risk: - Increasing provisioning coverage to commercial bank levels has limited impact. - Sharp deterioration of 10 percent of loans (performing to nonperforming) could breach minimum capital ratios for 30 percent of cooperative banking sector. - Concentration risk: - Simultaneous default of 5 largest borrowers for each cooperative bank could breach minimum capital ratios for 61 percent of cooperative banking sector (5 percent of total banking system assets). ### Regression analysis on bank profitability - Context: - Polish banks relatively more profitable compared with many European banks, but return on equity slightly below cost of equity. - Return on equity (banking system) = 10.4 percent. - EU average return on equity = 6.1 percent. - Cost of equity for Polish banks quoted in Warsaw Stock Exchange estimated around 11 percent. - Regression approach: - Panel regression controlling for bank characteristics, business models, industry structures, macroeconomics, cyclical factors, and monetary policy (Annex VII). - Dependent variables: ROAE and ROAA. - Sample: 34 commercial banks included in solvency stress tests. - Key regression findings: - Profitability driven by combination of structural and cyclical factors, including efficiency and interest rates. - Cost-to-income ratio negatively related to profitability. - Excess capacity (number of bank branches per capita) negatively associated with profitability. - Branch density in Poland relatively high in cross-country comparison (Annex I). - Short-term interest rate positively related to profitability (in particular ROAA); low interest rate environment puts downward pressure on bank profitability. *Source: IMF Staff Calculations; NBP.* ### 41. Short-term and long-term liquidity stress tests were conducted using Basel III ### 1polea2019005 - 41. Short-term and long-term liquidity stress tests were conducted using Basel III standards ### Overview of liquidity stress-testing exercise - Three liquidity stress tests were envisioned: the liquidity coverage ratio (LCR), the net stable funding ratio (NSFR), and a net cash flows-based liquidity stress tests. Due to data limitations only the LCR and NSFR tests could be conducted. - Four LCR stress scenarios were considered: baseline (CRR-recommended weights), adverse retail, adverse wholesale, and combined (retail + wholesale with solvency-test haircuts). - Cut-off date for liquidity stress testing: December 2017. Minimum threshold applied to LCR: 100 percent (January 2018 binding level). ### A. Funding structure and risks - Main vulnerabilities identified: - Short maturity of deposits and concentration in less stable deposits. - Limited secondary market trading of ‘liquid’ assets. - Key statistics and structure (as of December 2017): - Loan-to-deposit ratio for commercial banks: about 102 percent and declining. - About 70 percent of deposits have maturities of less than 30 days. - Only 31 percent of funding is classified as stable (according to EBA reporting templates). - Among the funding sources, 67 percent of retail deposits are classified as stable (Figure 10). - Funding composition notes: - Loans are funded mostly through inexpensive household deposits in local currency. - Deposits are mostly retail and concentrated in short maturities with little wholesale funding of either secured or unsecured nature by financial or non-financial customers. ### B. LCR-based liquidity stress-test scenarios and methodology - Four scenarios: baseline Basel CRR, adverse retail, adverse wholesale, combined adverse retail and wholesale. - Baseline replicated banks’ self-reported stress tests using CRR recommended stress weights; differences arose because banks used different weights within CRR discretion. - Adverse scenario calibrations: - Retail event: higher run-off rates applied to stable deposits and other retail deposits, calibrated as the highest run-off rate observed in the period of analysis. - Wholesale event: higher run-off rates applied to operational and non-operational deposits; historical monthly outflow events around 15 percent, mostly in household deposits and in all currencies. - Combined event: higher run-off rates applied jointly to stable deposits, other retail deposits, operational and non-operational deposits; haircuts on liquid assets derived from adverse solvency scenario. - Standard CRR weights applied to inflow roll-off rates across all events. - Data sources and mapping: - Calculations used FINREP/COREP data templates (EBA ITS reporting templates C72, C73, C74 for LCR) aggregated and mapped to IMF templates. - Bank-by-bank supervisory data provided by the NBP. - Macroeconomic variables taken from IMF International Financial Statistics, IMF World Economic Outlook database and Haver Analytics. - Selected LCR assumption excerpts (as presented): - Central government assets: CRR weights (recommended) 100% — Used 97% in scenarios. - Stable deposits: CRR weights (recommended) 5% — Retail Event used 10% (in specified column); Wholesale/Combined as shown in Table 4. - Other retail deposits: CRR weights (recommended) 10% — Retail Event used 15%. - Deposits subject to higher outflows categories and operational/non-operational deposit treatment varied by event as summarized in Table 4. ### C. LCR stress-test results (commercial banks) - Baseline event: - System LCR: 149 percent. - Five banks had an LCR above 500 percent. - Most banks had LCR between 100 and 200 percent. - One medium-sized domestic bank failed marginally; liquidity shortfall in that bank was negligible. - Adverse scenarios aggregate impacts: - Retail event: - System LCR remained above 100 percent. - Two medium-sized foreign banks failed with a liquidity shortfall of 0.5 percent of 2017 GDP. - Wholesale event: - System LCR fell to 120 percent. - Event affected eight banks including two large banks with sizable operational deposits. - Liquidity shortfall in percent of 2017 GDP: 0.5 - Combined (extreme) event: - System LCR fell to 88 percent (not linear; combined effect larger than sum of separate effects). - Affected 11 banks including three large ones. - Total shortfall compared to 100 percent coverage: less than 3.6 percent of GDP. - Liquidity shortfall in percent of 2017 GDP: 3.6 - Bank counts and asset shares (selected figures reported): - Total banks tested: 34 (by size: Large 5, Medium 14, Small 15). - Under Event 1 (baseline=149%): Banks with LCR <100: 1 (Percent of commercial bank assets) 0.0 1.7 0.0 = 1.7 (total). - Under Event 2 (system LCR=120%): Banks with LCR <100: 3 (Percent of commercial bank assets) 0.0 9.5 0.0 = 9.5 (total). - Under Event 3 (system LCR=94%): Banks with LCR <100: 25 (Percent of commercial bank assets) 31.4 13.4 0.7 = 45.6 (total). - Under Event 4 (system LCR=88%): Banks with LCR <100: 36 (Percent of commercial bank assets) 40.5 14.9 0.9 = 56.4 (total). ### D. LCR results by currency - System LCR in domestic currency: - Remained above 100 percent in most stress scenarios, except the extreme combined case. - System LCR in foreign currency: - Banks struggled to meet LCR requirements in foreign currency in the adverse scenarios. - All banks failed each of the stress events in foreign currency. - Retail event had stronger adverse consequences on foreign-currency liquidity than the wholesale event. - Caution: Currency-by-currency LCR data did not reconcile with aggregate foreign-currency LCR data (possible reason: banks report positions only in major foreign currencies). ### E. Cooperative banks - NBP conducted top-down LCR stress tests on cooperative banks using same assumptions as for commercial banks; results were similar. - Cooperative banks showed higher LCR levels than commercial banks due to higher surplus of deposits over loans. - Many cooperative banks did not pass the stress test at individual level, but they were exempt from complying with individual LCR by the regulator. - Key vulnerability: affiliating banks’ role — cooperative banks required to deposit a minimum amount of up to 9 percent of non-financial clients’ deposits in affiliating banks. By end-2017: - Affiliating banks’ total interbank deposits equaled 29 percent of total non-financial client deposits in cooperative banks. - Interbank deposits financed as much as 71 percent of affiliating banks’ assets. ### F. NSFR-based liquidity stress testing - NSFR analysis constrained by data availability and quality; based on EBA ITS reporting templates C60a, C60b, and C61 for NSFR. - Data quality concerns: mission team results did not reconcile with PFSA survey data (to which FSAP team had no access). - Reported results: - System NSFR = 101.9 - Total banks tested: 34 (by size: Large 5, Medium 14, Small 15). - Banks with NSFR <100: 38 (Percent of commercial bank assets) 36.2 19.3 1.3 = 56.9 (total). - By ownership: Banks with NSFR <100: 9 (Percent of commercial bank assets) 32.3 24.6 = 56.9 (total). - Liquidity shortfall in percent of 2017 GDP: 6.0 - Interpretation: - Mission’s results suggested the system in aggregate barely passed the NSFR test, with almost half of banks not meeting 100 percent coverage. - Authorities’ calculation of NSFR using PFSA bank survey data was higher at around 120 percent. ### G. Contagion and systemic risk analysis - Approaches used: - Exposure-based analysis using claims data from NBP or BIS and the Espinosa-Vega and Sole approach. - Market-based analysis using the Diebold and Yilmaz approach. - Interbank exposures: - Interbank exposures in Poland are generally contained; exposures between other Monetary and Financial Institutions (MFIs) declined to about 3 percent of total assets at end-2017. - Among commercial banks, small banks have the most substantial asset-side interbank linkages, but exposures are contained regardless of bank size. - Credit union interlinkages: - PFSA data indicate nearly no direct exposures of commercial and cooperative banks to credit unions by end-2017. - NBP data show deposits in banks amounted to less than 5 percent of credit unions’ total assets by June. *Source: IMF staff estimates and NBP data as reported in the FSAP chapter excerpt.* ### 2017. As credit unions represent only about 0.4 percent of total financial system assets, their credit ### 1polea2019005 - 2017. As credit unions represent only about 0.4 percent of total financial system assets, their credit ### Interbank network simulations and methodology - The mission examined interbank connectedness using the Espinosa-Vega and Solé (2010) approach; the NBP carried out simulations using FSAP team support. - Dataset: pair-wise on and off-balance sheet exposures between all 34 commercial banks, 553 cooperative banks, and 352 foreign bank counterparties; simulations covered individual failures of 939 banks. - Data limitations and assumptions: - The dataset did not distinguish between secured and unsecured exposures; NBP assumed all exposures to be unsecured. - First round (credit shock): assumed a 100 percent loss-given-default rate. - Second round (funding shock): assumed 35 percent of lost funding could not be rolled over; affected banks sell liquid assets at a haircut of 50 percent. - Third round: combined credit and funding shocks. - Rationale: 100 percent loss-given-default justified by limited data on collateralization and likely high uncertainty about asset values in severe stress. ### Key simulation findings (commercial banks only) - Failures of large banks have the most significant impact when only commercial banks are considered. - No more than one additional failure was induced in any of the commercial-bank-only simulations. - Outward spillover (index of contagion) was contained around or below 4 percent of Tier 1 capital. - Small and medium-sized banks were the most severely affected as measured by the index of vulnerability. - Funding shocks drove most effects, reflecting dependence on funding from large banks among some small- and medium-sized banks within the same capital groups. ### Cooperative banks, credit unions, and combined simulations - Cooperative banking sector size: about 5 percent of total financial system assets. - When commercial and cooperative banks were both considered: - The average number of bank failures induced by the medium-sized banks rose to 38. - The maximum number of induced failures rose to 338. - Results reflected failures of cooperative banks due to credit losses caused by affiliating bank defaults and structural features of the cooperative system. - Despite the rise in induced failures counts, the outward spillover of medium-sized banks increased only marginally (index of contagion) because cooperative banks are small relative to the system. - Only one simulation led to more than one round of induced failures (two rounds), indicating limited contagion when cooperative banks fail. - Note: the actual maximum value of cooperative banks’ index of vulnerability is 100 percent (figure axis capped at 5 percent in chart). ### Foreign counterparties - Simulations including 352 foreign counterparties: failures of only three foreign banks induced an additional failure. - In those cases, the average total assets of the additionally failed banks amounted to 0.5 percent of total system assets, indicating limited impact of foreign bank failures. ### Financial safety net channel - Banks and credit unions are required to contribute to the Bank Guarantee Fund (BGF) to ensure restructuring and deposit guarantee funds are sufficient. - In 2017 average contributions to these funds amounted to 0.1 percent of total commercial bank assets; contributions are not tax deductible and have a significant impact on profit and loss. - Collective contributions to restore the funds following bank failures represent a direct channel of contagion between institutions in the banking sector. ### Cross-sectoral interconnectedness and data constraints - Cross-sectoral analysis constrained by a lack of institution-specific data for non-banks; FSAP team relied on NBP Financial Stability Report, NBP Monetary Financial Statistics and PFSA data on total cross-sectoral exposures. - Sovereign-bank nexus material: - By end-2017, state-controlled banks represented as much as 40 percent of deposit-taking institutions’ total assets. - General government claims amounted to 20 percent of total assets of other MFIs by end-2017. - Foreign ownership linkages: - Foreign owners controlled 46 percent of banking sector assets and between 50–60 percent of insurance sector assets by June 2017. ### Sovereign-bank nexus (summary of channels and quantitative implications) - Channels: 1) sovereign exposure channel (holdings of domestic government securities by banks), 2) safety net channel (backstops, guarantees, direct state ownership), 3) macroeconomic channel (higher sovereign risk → contractionary effects). - Quantitative notes: - General government exposures of other MFIs amounted to 20 percent of their total assets by end-2017. - Prior to FIAT introduction in 2016 the exposure ratio was about 17 percent. - Under simple assumptions and based on current fair value holdings, a 500 bps increase in government yields reduces aggregate Tier 1 capital by about 2.2 percentage points. - Adjusting for the approximate 12 percent (relative) increase of sovereign debt holdings above trend following the FIAT introduction, the same yield shock reduces capital by 2.0 percentage points; a 0.2 percent lower impact than the current position. - By end-2017 the government directly or indirectly controlled 40 percent of the assets of deposit-taking institutions, up from 26 percent in 2015. ### Banks’ exposures to non-banks and non-bank dependencies - By end-2017 average large exposures of commercial banks to non-bank financial institutions amounted to 1 percent of Tier 1 capital; the largest exposure in the sample amounted to 25 percent of Tier 1 capital. - Total bank liabilities to the financial sector amounted to 14 percent of total assets, indicating limited dependency on financial sector funding. - Bank deposits accounted for 4–7 percent of total assets in the insurance and pension fund sectors by June 2017. - Pension funds had total exposure to the banking sector of about 41 percent of total assets by end-2017 (substantial holdings of bank equity). - Holdings of banking sector debt instruments made up 10 percent of open-ended funds’ total investments. ### Non-bank financial institutions and investment fund linkages - Insurance companies’ use of investment funds (June 2017): - Investment fund holdings in percent of total investments: 77 percent for unit-linked policies, 17 percent for life insurance policies, and 11 percent for non-life insurance policies. - Open-ended funds pose redemption/fire-sale risks, but use of open-end funds declined from 81 percent of total investment fund assets in 2010 to 49 percent in 2018, reflecting increased use of closed-end funds. - Unit-linked policies represented about 73 percent of total life insurance technical provisions by June 2017; risks associated with open-end funds are largely borne by policy holders rather than insurance companies. ### Market perception of interconnectedness (equity-based connectedness) - Methodology: Diebold and Yilmaz (2014) applied to daily equity log returns for 12 commercial banks and one insurance company from May 2015 to April 2018; VAR(3) with a predictive horizon of 12 days; rolling window of 100 days for total connectedness. - Market-based findings: - Group of large banks are net transmitters of shocks in the domestic financial system (outward spill-over from large banks to medium-sized banks and insurance companies tends to be higher than reverse). - Historical evolution of total connectedness has been quite volatile; average level of connectedness in the Polish system appears relatively contained. - The average level of domestic connectedness for Poland has been low compared with the Spain FSAP domestic index, Diebold and Yilmaz (2014) index, and the international index in the technical note. *Source: IMF staff and NBP simulations and analyses as presented in the cited FSAP materials.* ### 71. The FSAP analysis of cross-border interconnectedness was constrained by a lack of ### 1polea2019005 - 71. The FSAP analysis of cross-border interconnectedness was constrained by a lack of data ### Data constraints on cross-border interconnectedness analysis - Poland does not report data to the BIS International Financial Statistics database. - The BIS database only contained data on foreign banking system’s total exposures to Poland. - The FSAP team had access to an NBP dataset on bank-by-bank cross-border total exposures to other countries for Polish commercial banks. - Data constraints prevented: - a granular analysis of Polish banks’ credit exposures to foreign banking systems, - a granular analysis of Polish banks’ reliance on external funding. ### Aggregate external funding and asset-side exposures - Aggregate use of external funding by Polish banks has decreased significantly since the financial crisis. - This decline is largely due to the downsizing of Polish banks’ foreign currency housing loan portfolios, reducing the need for foreign currency funding. - According to the NBP Financial Stability Report almost all external liabilities were to financial institutions by June 2017; more than half of those liabilities were to related entities (intra-group funding of subsidiaries of foreign banks). - External assets of other MFIs amounted to 4 percent by March 2018. - Cross-border exposures (NBP dataset): - The average total cross-border exposure to one single country among the top-15 exposures of commercial banks was 2 percent of Tier 1 capital by end-2017. - The largest exposure to one country among the commercial banks was 93 percent of Tier 1 capital. - Implication: no bank fails due to a credit shock caused by the failure of another country using the Espinosa-Vega and Sole approach. ### Foreign banks’ exposures to Poland and market-based connectedness - Considering the group of 15 foreign banking systems with the largest exposures to Poland: - The average exposure amounted to 10 percent of Tier 1 capital by June 2017. - Portugal had the largest total exposure to Poland amounting to 58 percent of deposit-takers’ Tier 1 capital. - Likely driver: presence of foreign-owned subsidiaries in Poland (example given: Banco Comercial Português through Bank Millenium). - Implication: no foreign banking system would fail due to a credit shock caused by the failure of Poland using the Espinosa-Vega and Sole approach. - Diebold and Yilmaz methodology applied to two-day averages of daily log returns of MSCI equity indices for 19 counterpart countries: - The Polish banking system appears to be a net recipient of shocks according to market pricing (outward spill-over from foreign banking systems to Polish banks tends to be higher than the outward spill-over from Polish banks to foreign banking systems). - Poland was a net recipient of shocks in relation to almost all countries in the group. - Exception: the Polish banking system tended to be a net transmitter of shocks to the Portuguese banking system, consistent with Portugal’s significant cross-border exposures to Poland. - Note: two-day averages were used to account for time-zone differences. ### System-wide resilience, liquidity, and contagion findings - Systemic resilience: - The banking system in Poland shows resiliency to adverse shocks at the system level, however, some medium-sized banks including four OSIIs show weakness. - The main contributing factors to the decline in capital ratios under the adverse scenario are credit risks, funding interest rate risks and valuation risks on bond portfolios. - The necessary recapitalization need would be around 0.5 percent of GDP. - Four OSIIs, including the affiliating banks for cooperative banks, are found to be relatively weak. - Cooperative banking sector is exposed to credit and concentration risks (simulated default of the five largest borrowers). - Liquidity and funding: - The banks system appears resilient to short term liquidity risks in domestic currency. - At the aggregate level, banks were found to be resilient to the adverse retail event, but the wholesale and combined scenarios had a sizable impact on system liquidity. - The combined extreme event compounded costs that nevertheless totaled still less than 3.6 percent of GDP. - The LCR by currency showed resiliency in local currency but not in foreign currency. - NSFR test results appear to indicate vulnerability to more protracted events (subject to data uncertainty). - Contagion risks: - Contagion risks in the Polish financial system generally appear to be contained. - Adverse shocks to affiliating banks could spill-over and cause severe distress among cooperative banks, reinforcing the need to improve resilience of affiliating banks. - The sovereign-bank nexus in Poland is significant due to banks’ sizable holdings of sovereign securities and ownership linkages; this has become more material because of the FIAT that incentivized sovereign holdings. ### Policy recommendations (selected and verbatim where applicable) - Address the identified weakness in medium-sized banks and OSIIs, including the affiliating banks for cooperative banks, given their systemic importance in the banking system and strengthen the supervision and oversight of the relevant OSIIs and other weak banks. - Improve data for credit risk analysis, by collecting PD, LGD data and moving towards PD, LGD based credit risk models, in particular with the introduction of IFRS 9 that requires the estimations of expected losses and other forward-looking risk measures. - Improve monitoring of FX liquidity risks by ensuring LCR and NSFR currency-by-currency reporting. - Ensure banks’ readiness for the upcoming CRR II implementation timeframe. In addition, ensure that in practice banks are using common parameters for the LCR, to promote comparability across banks. - Replace the FIAT with a Financial Activities Tax (FAT), which would be less distortionary for the private sector and the economy than a tax on assets (IMF, 2010), addressing both stability and efficiency concerns. - Address data gaps in contagion analysis, by improving the collection of more granular data on: - counterparty-specific asset- and liability-side exposures and loss-absorbing capacity to improve the analyses of intra-sector and cross-sectoral contagion; - asset- and liability-side cross-border exposures to foreign banks; - banks’ secured and unsecured interbank exposures. *REPUBLIC OF POLAND, INTERNATIONAL MONETARY FUND* ### Annex II. Table 1. Poland: Risk Assessment Matrix (concluded) ### Annex II. Table 1. Poland: Risk Assessment Matrix (concluded) ### Domestic policy uncertainty and slippages - Overall Level of Concern: Medium - Likely outcomes and channels: - Inflation could accelerate faster than expected which would require more abrupt tightening. - Rapid tightening of interest rate could dampen domestic consumption and investment. - Fiscal slippages and the weakening of key institutions could result in rating downgrades and increase the perceived risk and uncertainties. - The interaction of fiscal and financial sector policy missteps could result in a vicious cycle of weaker public finances and financial sector health and lower growth. - Policy uncertainties surrounding the final solution to address consumer protection concerns associated with swiss franc mortgage loans and the potential cost could impact bank balance sheet and confidence. ### Domestic financial stress - Overall Level of Concern: High - Expected Impact and transmission: - Tighter domestic financial condition could dampen credit and increase NPL ratio and impair asset quality. - Tighter financial conditions both externally and domestically would hurt consumption and investment. - Rating downgrades could push up financing costs and banks’ funding costs. - Higher cost would then impact bank profitability and its ability to accumulate capital. - A rise in Polish sovereign bond yield could also lead to valuation losses for banks as they hold a sizable share of domestic government bond. - Potential costly solution to swiss franc mortgage loans could further impair bank profitability. *Note on RAM probabilities provided in source.* *Italic line omitted by pipeline.* --- ### Annex III. Cyber Risks - Monitoring and scope: - Cyber risks are monitored under financial institutions’ operational risk management framework in Poland. - Types analyzed include malware, phishing, Distributed Denial of Service (DDoS) attacks, advanced persistent threat (APT), data confidentiality and integrity breaches, among others. - For the central counterparty (CCP), maximum recovery time for the CCP’s critical functions shall not exceed two hours (legal requirement). - Practices by financial intermediaries: - Business impact analysis and stress tests used to assess potential cyber events. - Commercial bank: annual business impact analysis and stress tests focusing on APT and DDoS threats. - CCP: model-based approach for scenarios including hacking, theft or loss of data and breaches. - Business Continuity Plans and Disaster Recovery Procedures developed. - No dedicated cyber insurance purchased; cyber risks covered by other insurances targeting fraudulent activities and IT crimes. - Data collection and assessment: - NBP cooperates with ECB and PFSA to collect cyber-related self-assessments as part of the Cyber Resilience Strategy developed by the ECB. - PFSA collects information on security incidents and electronic banking via quarterly reports of key risk indicators from all commercial and cooperative banks. - Detailed information on number, type of cyber-attacks and incurred losses could only be collected with new domestic executive regulations. - Systemic assessment and quantified exposure: - At the current juncture, cyber risks are not considered systemic by the PFSA. - PFSA estimated potential loss exposure in electronic banking channel (due to lawsuits) to be about 26 million Zloty for commercial banks, and 1.6 million Zloty for cooperative banks in 2017, accounting for about 0.001 percent of GDP in total. - Losses from potential disruption of services or indirect reputational impact could be much higher. *Annex based on questionnaire responses from the NBP, the PFSA, and financial intermediaries including the CCP and one commercial bank.* --- ### Annex IV. Solvency Stress Test — Credit Risk Estimations - Data and methodology: - NPL ratios projected for individual banks for three exposure classes: corporate, mortgage and consumer. - Bank-by-bank quarterly data from 2008Q1 to 2017Q4 provided by NBP. - Macroeconomic series from IMF International Financial Statistics, IMF WEO, and Haver Analytics. - NPL ratios transformed using logistic transformation to capture non-linearities and ensure projections bounded between 0 and 1. - Panel regressions estimated with Arellano-Bover/Blundell-Bond linear dynamic panel-data estimator with robust standard errors. - Bayesian model averaging (BMA) used to guide variable and lag selection; regressors significant in BMA used in dynamic panel GMM estimations. - Transformation and forecasting formulas (as presented in source): - Logit transformation applied to original NPL ratios: ܻ௞,௝,௧ = ln(ܮܲܰ௞,௝,௧ / (1−ܮܲܰ௞,௝,௧)) - Dynamic panel specification (symbolic as in source): ܻ௞,௝,௧ = α + ρ ܻ௞,௝,௧−1 + β ܺ௧ + ܝ௞,௝,௧ - Inverse logistic function used to compute projected NPL ratios: ܮܲܰ௞,௝,௧ = 1 / (1 + exp(−ܻ௞,௝,௧)) - Key estimation findings (Annex IV. Table 1): - Lagged dependent variable coefficients: Corporate 0.899*** (0.12), Mortgage 0.883*** (0.04), Consumer 0.821*** (0.06). - Corporate: GDP growth (t-2) −0.0172* (0.01); Annual change in unemployment rate (t-1) 0.0280** (0.01); YoY Change in Equity Prices −0.00151* (0.00). - Mortgage: Annual change in unemployment rate (t-2) 0.0329* (0.02); Swiss Franc/Zloty exchange rate (t-2) 0.0863** (0.04); YoY Change in House Prices −0.00246** (0.00). - Consumer: Wage growth (t-2) −0.0100* (0.01); Long Term Interest Rates 0.0273*** (0.01). - Constants reported: Corporate −0.14 (0.24), Mortgage −0.700*** (0.20), Consumer −0.463*** (0.11). - Observations: Corporate 1,055; Mortgage 828; Consumer 1,092. - Number of Banks: Corporate 31; Mortgage 26; Consumer 30. - Note: Robust standard errors in parentheses; significance: *** p<0.01, ** p<0.05, * p<0.1. - Model implications: - NPL ratios are sensitive to GDP growth, changes in unemployment rates, exchange rates, wage growth and interest rates. - Corporate exposure: decline in real GDP and equity prices and a rise in unemployment could increase corporate default rate. - Mortgage exposure: mortgage default rates sensitive to unemployment rate, house price changes and Swiss Franc/Zloty exchange rate movements. - Consumer exposure: default rates associated with wage growth and long-term interest rates. --- ### Annex V. Solvency Stress Test — Funding Rate Estimations - Data and methodology: - Bank funding costs estimated using dynamic panel regression (Arellano-Bover/Blundell-Bond) with robust standard errors for 34 commercial banks. - Quarterly bank-by-bank data from 2008Q1 to 2017Q4 from NBP; macroeconomic series from IMF IFS and Haver Analytics. - Estimation results (Annex V. Table 1): - Dependent variable: Effective Interest Rate on Liabilities. - Lagged dependent variable 0.677*** (0.03). - 3-month Wibor (t-2) 0.224*** (0.03). - Long term interest rate (t-2) 0.0500*** (0.02). - Constant −0.111** (0.06). - Observations 1,194; Number of Banks 34. - Note: Robust standard errors in parentheses; significance: *** p<0.01, ** p<0.05, * p<0.1. - Transmission to lending rates: - Relationship between changes in funding rates and in lending rates in the adverse scenario estimated using quantile regressions (tau = 0.8). - Quantile regression results (Annex V. Table 2): Change in Effective Interest Rates on Liabilities coefficient 0.521*** (0.06); Constant −0.045* (0.03). - Adjusted R-squared: 0.505. Sample (adjusted): 2009Q4 to 2017Q4. - Imperfect pass-through implies tightening in interest rate margins for banks in the stress testing scenario. --- ### Annex VI. Background Risk Analysis of FX Mortgages - Historical context and size: - FX mortgages mostly in Swiss franc became popular prior to the Global Financial Crisis. - FX mortgage portfolio grew from 2 percent of GDP (around 60 percent of total outstanding housing loans) at end-2004 to 11 percent of GDP (around 70 percent of total outstanding housing loans) by end-2008. - At peak, medium-sized and foreign-controlled banks (incorporated in Poland) were most exposed; borrowers tended to be wealthier households. - Performance and resilience: - FX mortgages had relatively low NPL ratios in the last decade (below 4 percent), even during stress episodes. - During 2015 Swiss franc appreciation, NPL ratios of this portfolio rose only moderately. - External mitigating factor in 2015: simultaneous reduction in Swiss LIBOR rate partially offset increased debt service from Swiss franc appreciation. - Domestic mitigating factor: prudential regulations tightened since 2006—measures include stricter risk assessment, buffers for banks and borrowers, limits on borrowers’ risk exposure, and effective ban on new FX mortgage loans to unhedged borrowers from 2013. - Stock of FX mortgages declined from early 2012 with an average annual amortization rate of 8 percent in recent years, reaching 6½ percent of GDP by end-2017. - Deterioration in risk profile and public debate: - Average LTV ratios for FX mortgage loans increased considerably after the 2015 Swiss franc appreciation, especially for loans granted during 2007–08. - By end-March 2015, share of Swiss franc mortgages with LTV higher than 80 and 100 percent reached 68 and 53 percent of total Swiss franc mortgages outstanding (based on PFSA survey data including banks with FX loans over 10 percent of total loans). - Debt service burden on Swiss franc mortgages rose considerably. - NBP concluded after March 2015 that FX mortgage portfolio posed no systemic risk to banking sector stability, though public concerns increased and calls for borrower protections intensified. - Legislative proposals and regulatory responses: - Four draft bills under Parliamentary discussion (two involve mandatory conversion submitted by Civic Platform and Kukiz’15). - August 2016 presidential proposal: require banks to repay “excessive” FX spreads; assessed to cost between 4 to 9 billion zloty (up to three-quarters of the 2016 banking sector profits). - August 2017 presidential proposal: establish “Restructuring Fund” with quarterly contributions from banks (capped at 0.5 percent of the FX mortgage stock) to reimburse banks for costs incurred during FX mortgage restructuring; contribution capped at 2¾ billion zloty for the first year with current stock of FX mortgages. - FSAP notes cumulative cost difficult to estimate due to uncertain speed of conversion and variable per-conversion cost. - FSC-M released nine recommendations in early 2017 including: increase risk weight for FX mortgages to 150 percent; impose a 3 percent systemic risk buffer (from zero percent) regardless of banks’ FX mortgage exposure; require additional capital for banks with FX mortgage portfolios. Most recommendations in effect since late 2017 or early 2018. - Progress in voluntary FX mortgage conversion has been limited; borrowers have little incentive to convert given substantial interest rate differential. - FSAP assessment: - Risks associated with FX mortgage portfolio have declined and no systemic policy response is warranted. - Improvement supported by robust wage growth, favorable Swiss franc interest rates, income buffers of FX borrowers, and tightened prudential regulations following 2017 FSC recommendations. - FSAP stress tests show that impact of significant exchange rate depreciation on banking sector solvency is generally contained in aggregate, though pockets of vulnerability exist; vulnerabilities should be addressed case by case. - Scenario analysis for a potential “Restructuring Fund” (assumptions and approach): - Two scenarios provide bounds on cumulative cost if Restructuring Fund established: “no conversion” (upper bound) and “fast conversion” (lower bound), assuming contributions set at maximum level throughout scenario period. - “No conversion” scenario: banks pay contributions proportional to gradually amortizing stock (upper bound). - “Fast conversion” scenario: stock reduces more rapidly due to conversion, reducing contributions paid by banks. - Key assumptions: - Haircut rates that equalize pre- and post-conversion debt service (in zloty terms) would be acceptable to borrowers. - Cumulative debt relief a bank is willing to offer each quarter is equal to its contribution. - Only FX mortgages “at risk” (i.e., with LTV above 80 percent and DSTI above 50 percent) would be converted. - Costs calculated for both the main adverse scenario used in the solvency stress test, and the 30 percent zloty depreciation scenario in the sensitivity analysis. - Text from source (concluding fragment as provided): - "Under the main adverse scenario, the three-year cumulative cost ranges between" *Italicized source attribution line.* ### 7.3 and 8.1 billion zloty (0.2–0.3 percent of RWAs annually). The cumulative amount “at ### 1polea2019005 - 7.3 and 8.1 billion zloty (0.2–0.3 percent of RWAs annually). The cumulative amount “at ### FX mortgage portfolio: stress scenarios and implications - Under baseline/ongoing conversion assumptions: - The cumulative amount “at risk” after the shocks is around 26 percent of the end-2017 stock of FX mortgages, which could take 4½ years to convert. - Significant haircuts (25–30 percent) are needed to equalize monthly installments pre- and post-conversion. - The cost (relative to RWAs) is the highest for larger medium-sized and small banks, as well as for foreign-controlled banks. - Under the 30 percent depreciation scenario: - The one-year cost ranges between 3 and 3.3 billion zloty (22–25 percent of net profits and other comprehensive income). - The total amount “at risk” after the one-time shock is around 75 percent of the end-2017 stock of FX mortgages, which could take 11½ years to convert. - A haircut rate of 29 percent is needed to equalize monthly installments pre- and post-conversion. - Policy conclusion: - Risks associated with the FX mortgage portfolio has declined and no systemic policy response is needed. - Solutions to distressed FX mortgages should be case-by-case and risk-based. - If any policy actions are to be taken to meet objectives other than addressing financial stability risks, they should aim to minimize the adverse impact on bank profitability and financial stability risks. - Note on interpretation: - Bank-level cost distribution should be interpreted with caution, as the risk profile of FX mortgages is assumed to be the same across banks due to data limitation. - For more details on the assumptions and calculations, see Box 3 of the Background note on the “Risk Analysis of Foreign-currency Mortgages.” ### Regression analysis on bank profitability (Annex VII) - Sample and data: - Sample of 34 commercial banks in Poland, as in the solvency and liquidity stress tests. - Bank-by-bank balance sheet data from the NBP. - Macroeconomic variables from IMF International Financial Statistics, IMF World Economic Outlook database and Haver Analytics. - Methodology: - Panel data techniques using the Arellano-Bover/Blundell-Bond linear dynamic panel-data estimator with robust standard errors. - Dependent variables: headline profitability measures including ROAE and ROAA. - Model includes bank-fixed effects, lagged dependent variable, bank-specific indicators, banking industry structural indicators, and macroeconomic/cyclical variables. - Bank characteristics and proxies: - Bank size: natural logarithm of total assets. - Credit risk: loss provisioning ratio. - Cost efficiency: cost-to-income ratio. - Excess capacity: number of bank branches per capita. - Business model: share of non-interest income over total revenue; deposit to total assets ratio. - Key empirical findings (Annex VII. Table 1): - Lagged Dependent Variable: 0.813*** (ROAA), 0.905*** (ROAE) with robust standard errors (0.01) and (0.03) respectively. - Average Assets (ln): 0.096 (ROAA) and -0.439 (ROAE) with standard errors (0.13) and (1.04). - Cost to Income ratio: -0.00650*** (ROAA) and -0.0459*** (ROAE) with standard errors (0.00) and (0.01). - Non-Interest Income to Revenue Ratio: -0.0572 (ROAA) and 0.362 (ROAE) with standard errors (0.07) and (0.23). - Bank Branch per capita: -0.0400* (ROAA) and -0.163** (ROAE) with standard errors (0.02) and (0.08). - 3 month Wibor: 0.0551* (ROAA) and 0.109 (ROAE) with standard errors (0.03) and (0.28). - Real GDP growth: -0.0340 (ROAA) and 0.116 (ROAE) with standard errors (0.03) and (0.11). - Constant: -1.8481 (ROAA) and 0.91 (ROAE) with standard errors (2.87) and (24.58). - Observations: 1190 (for both ROAA and ROAE). - Number of Banks: 34. - Note: Robust standard errors in parentheses; significance markers: *** p<0.01, ** p<0.05, * p<0.1. All independent variables are lagged by one quarter. - Interpretation: - Cost to income ratio is negatively related to profitability. - Excess capacity (branches per capita) is negatively associated with profitability. - Short-term interest rate is positively related to profitability (in particular ROAA). - Deposit to asset ratio and the CR5 index are found to be insignificant in most cases. ### Contagion analysis methodology (Annex VIII) - Two complementary approaches used: - Espinosa-Vega and Solé (2010) methodology for contagion through balance sheet exposures (network analysis). - Diebold and Yilmaz (2014) methodology for contagion according to market perceptions (market-data VAR and Generalized Variance Decomposition). - Network analysis (Espinosa-Vega and Solé, 2010): - Considers both credit and funding shocks. - Credit shock: conservative loss-given-default rate of 100 percent assumed. - Funding shock: bank B assumed to lose 35 percent of funding from bank A, replaced by selling liquid assets at a 50 percent haircut. - Cross-border analysis only considered the credit shock due to data limitations. - A “failure” is assumed when incurred losses exceed Tier 1 capital. - Additional rounds of contagion (domino effects) allowed in interbank network analysis. - Contagion analysis with market data (Diebold and Yilmaz, 2014): - Estimates a VAR with market data; derives interconnectedness from Generalized Variance Decomposition (GVD). - GVD measures fraction of changes in one variable attributable to shocks to another variable. - Two sets of estimations: interconnectedness among publicly traded banks and one insurer in Poland; interconnectedness among banking sector equity indices in 19 countries. - Uses daily log equity and index returns; average two-day log returns used for cross-border analysis to control for trading-hour differences. - Constructs pair-wise directional connectedness, total connectedness, and net contribution measures. ### Stress testing frameworks and matrices (Annex IX) - Solvency stress test matrix (using December 2017 data): - Institutions included: 34 commercial banks and 553 cooperative banks. - Market share coverage: 85 percent of total banking sector assets (commercial banks); 7 percent of total banking sector assets (cooperative banks). - Data and baseline date: FINREP/COREP and supervisory data from NBP; December 2017. - Stress test horizon: Three-year horizon: 2018–20 for FSAP top-down; instantaneous sensitivity analysis for NBP top-down sensitivity analysis. - Tail shocks: - Macroeconomic scenario: baseline based on December 2017 WEO projections; adverse scenario with external financial market stress, growth slowdown, tightened financial conditions, and domestic policy slippages. - Sensitivity tests include additional exchange rate shocks (30 percent Zloty depreciation); interest rate shocks; counterparty risk shocks (simultaneous default of the five largest non-financial borrowers). - Behavioral adjustments: balance sheet grows in line with nominal GDP with a floor at zero; dividends are paid out by banks that remain adequately capitalized with positive profits; invariant asset allocation over three years. - Calibration: based on credit models estimated by IMF staff; CRD IV / CRR fully loaded levels for CET1; capital shortfalls measured in terms of CET1, T1, total capital and leverage ratio. - Output presentation: system-wide capital shortfall; number of banks and percentage of banking assets falling below certain ratios. - Liquidity stress test matrix (using December 2016/2017 data): - Institutions: 34 commercial banks and 553 cooperative banks. - Methodology: extended Basel III LCR scenario with variants (retail/wholesale/combined shock). - Risks and buffers: funding liquidity and market liquidity; buffers include counterbalancing capacity and central bank facilities. - Tail shocks: run-off rates calculated following historical events and LCR/NSFR rates; bank run and dry up of wholesale funding markets with haircuts to liquid assets. - Regulatory standards: Basel III liquidity standards for LCR and NSFR. - Reporting: liquidity ratios disaggregated by type (domestic vs. foreign) and size of bank. - Interconnectedness analysis matrix: - Cross-border analysis using Espinosa-Vega and Solé (2010) with BIS consolidated statistics, NBP data and regulatory capital data as of 2017; list of economies considered includes France, United Kingdom, Germany, Czech Republic, Luxembourg, United States, Netherlands, Belgium, Italy, Ireland, Sweden, Austria, Switzerland, Portugal, Spain, Japan, Russia, South Korea, Greece, Australia, Canada, Taiwan Province of China, Finland, Turkey, Chile, Norway, Slovakia, Ukraine, and Cyprus. - Diebold and Yilmaz (2014) methodology for cross-border interconnectedness with a list of countries to be considered: France, United Kingdom, Germany, Czech Republic, United States, Netherlands, Belgium, Italy, Ireland, Sweden, Austria, Switzerland, Portugal, Spain, Japan, Norway, Slovenia, and Russia. - Interbank market analysis by NBP using Espinosa-Vega and Solé (2010) with three simulations: - Simulation 1: contagion of each bank’s failure. - Simulation 2: contagion of collective failure of banks failing the solvency stress test; two thresholds tested: (i) 6 percent Tier 1 capital ratio and (ii) 3.5 percent Tier 1 capital ratio. - Simulation 3: contagion due to each foreign bank exposure’s failure. - Cross-sector analysis using Diebold and Yilmaz (2014) to examine spillover risks among publicly listed Polish banks and one insurance company using daily equity returns from 15/08/2015 to May 2018 and robustness checks with longer series. *Source: https://www.imf.org/-/media/files/publications/cr/2019/1polea2019005.pdf* --- _Source: https://www.imf.org/-/media/files/publications/cr/2019/1polea2019005.pdf_