## _cr1416 ## Source details **Canonical URL:** [_cr1416](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2014/_cr1416.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2014/_cr1416.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2014/_cr1416.pdf.json) --- ### EXECUTIVE SUMMARY — INTRODUCTION AND CONTEXT - Austrian banking system in a recovery phase following the 2008–2009 global financial crisis. - Three mid-sized domestic banks were fully or partially nationalized; Financial Market Stability Act (FinStaG) authorized recapitalization up to €15 billion; as of Dec 2012, €13.6 billion were utilized, of which capital injections reached €5.9 billion. - Consolidated banking sector assets about €1.1 trillion, or more than 3.5 times GDP in 2012 Q3. - Three large internationally active banks account for almost half of total bank assets. - Bank business models: two broad groups each account for about 45 percent of banking system assets; (partially) nationalized medium-sized banks represent about 7 percent of total assets. - Net interest income is the primary profit source; in 2012 net interest income amounted to almost three times the income from securities holdings. Return on assets stood at 0.3 percent in 2012. ### MACROECONOMIC OUTLOOK USED IN TESTS - Baseline forecasts indicate a gradual macroeconomic upturn; CESEE medium-term growth prospects stronger than advanced economies. - CESEE baseline: BIS-weighted average of individual countries’ projections as of June 2012. - WEO projections used in the stress test: as of Oct 2012. - Example macro projection for Austria: 2-year cumulative growth over 2013-2014 is projected at 3.05 percent under the baseline and at -2.97 percent under the adverse scenario. ### SOLVENCY STRESS TESTS — MAIN RESULTS AND CAVEATS - Aggregate capital buffers sufficient to withstand severe but plausible macro shocks. - Under the most severe scenario, estimated total capital shortfall amounts to 1 percent of GDP. - Three-year stress-testing horizon (2013–2015) does not consider: - Repayment of state participation capital grandfathered under Basel III until 2018. - Potential capital surcharge on domestic systemic institutions (from 2016). - Basel III positioning: - Aggregate sector would comfortably pass Basel III phase-in CET1 hurdle rates under the most severe scenario. - CET1 impact from front-loading full Basel III: sector CET1 would shed 1.4 percentage points relative to EBA CT1; large international banks 1.6 percentage points. - Impact on Tier 1: 1.7 percentage points for the system and 2.2 percentage points for the large international banks. - CAR impact depends on treatment of €19.8 billion (€10.0 billion) long-term subordinated debt; CAR effect could range between 1.9 and 4.0 (2.3 and 4.9) percentage points for the banking system (large international banks). - Under the most severe macroeconomic scenario: - Banks representing less than 7 percent of total bank assets would fall below the regulatory threshold. - Estimated aggregate capital needed to restore these banks above regulatory minimum: 0.3 percent of total bank assets, or about 1 percent of GDP. - Percent of total bank assets in the 6–8 percent core Tier 1 bucket increases from about 4.5 percent (baseline) to 17 percent (severe scenario) at end of stress horizon. - Projected losses dent large banks’ core Tier 1 ratio by 4.8 percentage points versus average 2.8 percentage point losses for the Austrian banking system. - Drivers of capital decline: - Credit risk in CESEE is the main driver: more than two-thirds of expected losses from credit risk come from CESEE exposures; under one-fourth originate domestically. - Projected profit and loss effect from severe macroeconomic stress adds up to 1.5 percentage points. - Indirect credit risk from FX appreciation in foreign currency loans adds up to 35 bps in CT1 ratios for the whole banking sector. - Market risk from underperformance of market instruments associated to RPVs contributes an additional 7 bps. - Securitization exposure risk estimated impact: 10 bps. ### LIQUIDITY STRESS TESTS — MAIN RESULTS - Liquidity stress-test scope: largest 29 banking institutions (~80 percent of total assets). - Severe 30-day funding stress scenario: total liquidity shortfall estimated at 0.1 percent of total liabilities. - Aggregate liquidity shortfall results under severe assumptions: - 30 day horizon: 0.1 percent of total liabilities. - 90 day horizon: 0.3 percent of total liabilities. - 1 year scenario: 0.2 percent of total liabilities. - Time horizons: 30-day, 90-day, one-year; scenarios: baseline, market mild, market medium, market severe, combined. - Counterbalancing capacity approaches include Full counterbalancing capacity, Increased focus on market liquidity, and Market liquidity (haircuts up to 100 percent for securities < A-). - Foreign currency liquidity: substantial improvement in USD and CHF since 2008; CHF funding: half of banks made substantial progress; the other half need to diversify funding sources and lengthen funding tenors. ### CROSS-BORDER LINKAGES AND EXPOSURES - Direct and cross-border lending exposures nearly €460 billion. - €326 billion exposures to CESEE countries or under 30 percent of overall banking system assets. - Total assets of foreign subsidiaries in CESEE reached €234 billion, or over 70 percent of total exposure to the region, in Sep 2012. - Over 80 percent of aggregate subsidiary assets in CESEE are concentrated in the large internationally active banks. - Austrian banks’ market shares in CESEE: above one-third in Slovakia, Bosnia, Romania, Albania, and the Czech Republic. - Exposure to European peripheral countries fell to €31 billion from €45 billion in 2008; of the €31 billion, 17 percent were claims on the public sector and 24 percent claims on credit institutions. ### FUNDING/NETWORK ANALYSIS — METHODOLOGY AND FINDINGS - Funding/network model uses Austrian bilateral exposure matrix and default cascade (Furfine) model; LGD on net exposures assumed 100 percent; a bank is defaulted when consolidated CAR < 8 percent. - One-year global funding scenario assumptions include closure of unsecured interbank, FX swap markets and capital markets, contingent liability rise from 50 percent jump in drawings of committed lines, increases in funding costs (expiring term retail funding +140 bps; new issuance +70 bps for those able to access markets). - Fire sales mechanics: banks with negative funding gaps engage in distress sales; recovery rates of creditor losses = 0 percent. - Key findings: - System resilience: Austrian system resilient to funding and contagion stress transmitted through Austrian interbank market. - Aggregate impact: impact on core Tier I ratio of whole system limited to 108 bps, driven primarily by fire sales rather than funding cost effects or cascading defaults. - Caveats: fire-sale assets calibrated exogenously; spiral effects from aggregate supply not modeled; mark-to-market effects from common exposures not computed; contagion outside Austrian interbank market not considered. ### CONTAGION THROUGH MARKET CHANNELS (CoVaR) — METHODOLOGY AND FINDINGS - CoVaR analysis uses weekly market-valued asset returns over Apr 2005–Dec 2012; two peer groups: European banking system (40 internationally active banks) and CESEE banking system (14 large foreign banks active in CESEE). - Predictive state variables: liquidity strains have strongest predictive power; equity volatility, short-end yield movements, and sovereign debt crisis also relevant. - Asymmetry: individual bank distress impact on system tail returns is asymmetric — for European system the coefficient of negative returns on banking system returns is > four times larger than for positive returns; for CESEE peer group it is > two times larger. - Cross-border spillovers: - Outward spillovers from top two Austrian banks to CESEE banks are not negligible but smaller than systemic risk introduced by severe distress affecting other CESEE peer banks. - Inward spillovers to Austrian banks are most pronounced from CESEE-headquartered banks active across CESEE or from foreign banks with highest presence in the region. - Micro vs. macro prudential implication: weak link between individual solvency risk (VaR) and contribution to systemic risk (∆CoVaR); policy calibrations for capital add-ons should draw on both individual and systemic risk analysis. ### CREDIT RISK MODELING APPROACH — DOMESTIC AND CROSS-BORDER - EAD split: domestic 56 percent; cross-border 44 percent. CESEE exposures ~70 percent of cross-border exposures. - Domestic exposures: - OeNB credit risk model links sectoral corporate PDs (six sectors: construction, production, trade, transport, tourism and services) to nineteen macroeconomic time series and an unobserved component (latent risk factor); estimated over 1985–2011 with up to six quarterly lags. - Model selection uses Forward-Stepwise Selection; for models with >7 macro variables, latent factor significance vanishes. - Cross-border operations and subsidiaries: - Satellite model based on country-specific CESEE loan loss provisioning ratios (stock and flow) calibrated to proxy PD changes. - Stressed LGD: - Separate for collateralized and uncollateralized exposures. - Real estate collateral: country-specific haircuts estimated for CESEE based on elasticity of GDP growth to house prices. - Uncollateralized exposures: country-specific LGD capped at 45 percent for Austria; linked to World Bank Doing Business Statistics for CESEE; distribution truncated at 80 percent; stressed via linear quarterly increments reaching a final add-on of 10 percentage points in 2015Q4 under adverse scenario. - Profit dynamics and FX depreciation: - Assumed sustained cumulative depreciation reaches 1.5 (adverse) and 2.0 (severe) SD under 2013–2014, with partial rebound in last year. - CESEE subsidiaries’ share in consolidated net operating profits: over 50 percent in 2012 Q2; share in consolidated assets: below 25 percent in 2012 Q2. - Operating profits assumed to decline in line with net interest income. ### FOREIGN CURRENCY LENDING (FCL) AND RPV SENSITIVITIES - Domestic FCL stock (2012 Q3): €50.7 billion = 15.3 percent of all domestic loans; households €34.6 billion (25 percent of housing loans); corporates €10.0 billion (7 percent of loans to non-financial corporates). - Loan structure: - ~60 percent of FCLs to households and corporates are bullet loans with repayment vehicles (RPVs); 40 percent are amortizing. - Of €30.0 billion bullet loans outstanding in Sep 2012, €25.8 billion were FCLs, of which €24.0 billion were denominated in CHF. - 60 percent of all FCLs are linked to RPVs. - As of 2012Q3, almost 80 percent of FCLs associated to a RPV had maturity beyond 2020. - Provisioning rate on FCLs to non-banks: 1.1 percent in 2012 Q3 (less than one third of provisioning for euro loans). - Swiss Franc FX sensitivity scenario: - Assumption: CHF appreciates by 1.5 SD over 2012Q3–2014Q4; nominal exchange rate climbs from 1.21 to 1.07 and stabilizes thereafter — equivalent to CHF appreciation of 13 percent over 2012Q3–2014Q4. - Indirect credit risk elasticity: ΔLLPR = −5.2 * (ΔD* / I) (as presented in source). - Additional impairments distributed equally over each loan’s remaining maturity; stress test loss is accumulated loss over 2012Q4-2013Q1. - RPV statistics (Q3 2012): - 11.2 percent of net loans to private sector were RPV-based. - RPV loans to domestic nonbanks: €30.0 billion — €27.3 billion to households (20.7 percent share), €2.7 billion to non-financial corporations (2.0 percent share). - Mid-2011 total funding gap of RPVs: €5.3 billion (18.2 percent). - Mid-2011 FC-denominated RPV funding gaps: €4.7 billion (CHF and JPY funding gaps 20.1 percent and 18.6 percent respectively). - Mutual funds-based life insurance account for 54.0 percent of RPVs and have funding gap 21.0 percent. - RPV stress sensitivity: - Adding yield shocks (100 bps adverse, 200 bps severe) combined with 1.5 SD CHF appreciation increases maximum funding gap from accumulated €4.2 billion to €8.6 billion. - Given average remaining maturity of 13 years, implies additional loss under €1 billion during the stress test horizon. ### SECURITIZATION, SOVEREIGN AND MARKET RISK TREATMENT - Securitization: stress applied via increase in RWAs using credit risk migration matrix in line with 2011 EBA calibration; defaulted exposures risk-weighted at 1,250 percent. - Sovereign risk: - Scope: all central governments (but no central banks), all regional governments, all local authorities; CESEE subsidiaries’ sovereign holdings included. - Calibration: term structure shifts upward; for fifty countries shock derived from 90th percentile of historical annual changes of daily yields (Bloomberg 5-year yields 2005–12); for Belarus, Bulgaria, Luxembourg, Malta, Romania use extreme returns for most liquid international bond as of Dec 2012; Cyprus calibrated from sovereign yield curve as of Dec 2012. - Haircuts to banking book applied to adjusted (marked-to-market) balance sheet values; valuation loss = amortized cost - market value + market value * haircut. - Market risk: - Bottom-up (five largest banks) market risk sensitivity using trading book positions as of 31 Dec 2012; volatilities scaled by factor of two and over one-year horizon; FX and rate baskets and volatilities follow EBA 2011 calibration for emerging and major currencies. - Selected asset shocks preserved verbatim: Gold/USD%-20; Equity Europe (Eursostoxx 50)%-15; US (S&P 500)%-10; Japan (Nikkei)%-20; Ems (MSCI)%-20. ### LIQUIDITY STRESS TEST PARAMETERS (SELECTED, PRESERVED VERBATIM) - Market risk parameters (selected verbatim): - Interest Rates — Advanced Countries: USD 3Mbp10; USD 2Ybp50; USD 10Ybp150; EUR 3Mbp30; EUR 2Ybp70; EUR 10Ybp130; UK 3Mbp20; UK 2Ybp90; UK 10Ybp170; Other non-Emerging Markets 3Mbp60; Other non-Emerging Markets 2Ybp100; Other non-Emerging Markets 10Ybp110. - Emerging Markets: Eastern Europe 3Mbp100; Eastern Europe 2Ybp110; Eastern Europe 10Ybp140; Asia 3Mbp40; Asia 2Ybp40; Asia 10Ybp50. - Volatilities: US Volatility%40; EUR Volatility%40; UK Volatility%40; Other non-Emerging Markets Volatility%70; Emerging Markets Volatility%70. - Foreign Exchange FX Rates: USD/EUR%18; JPY/EUR%25; GBP/EUR%12; Other non-Emerging Markets currencies/EU%17; Emerging Markets currencies/EUR%15. - Liquidity Stress Test Medium Scenario Parameters (selected verbatim): - Cash-Inflows: Complete dry-up of unsecured interbank market (100%); (Complete) dry-up of FX-swap market (30/90 days: 100%; 1 year: 20%); Issuance of long-term unsecured bonds (30/90 days: -50%; 1 year: -20%); Issuance of short-term unsecured paper (30/90 days: -50%; 1 year: -20%); Issuance of long-term covered bonds (30/90 days: -50%; 1 year: -20%); Issuance of short-term secured paper (30/90 days: -50%; 1 year: -20%); Reduction of inflows from maturing loans & from paper in own portfolio maturing (based on a link to the solvency stress test). - Cash-Outflows: 30/90 day horizons: Wholesale and retail deposits (bank individual values based on conservative assessment, average across banks -2%); An additional idyosincratic shock reduces expected rollover rates of wholesale deposits to 90% and retail deposits to 95% over the 30 day period; 1 year horizon: wholesale and retail outflows calibrated on peripheral Euro area experience during sovereign debt crisis (wholesale -3%, retail -2%); Draw down of committed lines by other banks and non-banks (+50%). - Counter-balancing capacity: Haircuts on non-marketable unencumbered collateral deposited at OeNB increase to 100%; Haircuts on collateral deposited at OeNB with rating below A- increase to 100%; Haircuts on unsecured issuances by banks and financial corporates as well as Asset Backed Securities deposited at the OeNB increase to 100%; Haircuts on unencumbered collateral in USD, CHF, JPY, GBP, Other Currencies deposited at central banks (after CB haircuts) increase by additional 500 bp; Haircuts on marketable collateral AAA to A- (not pledged at central banks) increase by 100 bp, 500 bp, and 1000 bp, respectively; Haircuts on committed liquidity lines received by respective banks and liquidity injections from parent bank increase to 100%; Banks behavioural reactions are factored into the 1 year scenario (reduction of unsecured interbank loans granted -100%, reinvestment of maturing liquid assets -50%, new loans granted -3%); Haircuts on unencumbered marketable collateral with rating from AAA to A- pledged at OeNB increase by between 100 bp (asset category 1, credit quality step 1) and 1500 bp (asset category 3 and credit quality step 2) on top of OeNB standard haircuts; other asset categories and credit quality step collateral receive 100% haircuts; Haircuts on BBB rated (or below) non-financial corporate bonds, other marketable, pledgable (but not pledged assets) collateral, and other non-marketable pledgeable collateral (but not pledged) increase to 100%. ### POLICY RECOMMENDATIONS AND PRIORITIES (HIGHLIGHTS) - High priority, near-term: - Assess impact of different regulatory ratios across CESEE jurisdictions and potential ring-fencing of cross-border flows on consolidated regulatory ratios of Austrian banks. - Continue developing funding/contagion analysis by endogenizing fire sales and funding costs; allow domino effects from cross-guarantee schemes and cross-holdings of unsecured paper; extend bilateral exposure networks to global banks. - High priority, medium-term: - Consider modeling credit risk in CESEE based on insolvency data—subject to data availability—to avoid reliance on loan loss provisioning amid asset-quality concerns. - Medium priority, immediate to medium-term: - Include market risk factors, including rises in risk premia, in the satellite model for credit risk in domestic and cross-border exposures. - Develop framework to identify and measure banks’ individual contribution to systemic risk drawing on market-based approaches such as CoVaR. - Integrate solvency, liquidity, and market shocks in stress test scenario design, include second-round effects—particularly on credit growth—and allow for transmission of behavioral shocks. - Supervisory and bank actions: - Banks should continue building capital buffers gradually to support market confidence. - Banks should continue lengthening funding tenors in foreign exchange. ### RISK ASSESSMENT MATRIX — SELECTED ENTRIES - 1. Global shock/Intensification of the Euro Area debt crisis - Overall Level of Concern: Medium - Likelihood in next 1–3 years: Low/Medium (IMF Global Risk Assessment Matrix) - Expected Impact on Financial Stability if realized: High - 2. Contagion to/from CESEE countries - Overall Level of Concern: Medium - Likelihood in next 1–3 years: Medium - Expected Impact: High - 3. Severe Funding Stress of Global Banks - Overall Level of Concern: Medium - Likelihood in next 1–3 years: Medium - Expected Impact: Medium ### ANNEX HIGHLIGHTS — MARKET-BASED ANALYSIS, ECONOMETRICS, SOVEREIGN HAIRCUTS - Annex II market-based approach: - Time series: Oct 2007 to Oct 2012; regress monthly changes in CDS spreads and EDFs on Austria macro-financial variables, CESEE contagion proxies, and global risk factors including equity premium, volatility premium, and term premium. - Key empirical findings: - Upward revision of unemployment forecast is main Austrian macro determinant of solvency risk changes. - Contagion from CESEE is a notable risk factor; explanatory power increases from 0.1 to 0.5–0.6 when included. - Heightened stress in European banks (itraxx senior financial) and sub-investment grade corporates widen solvency risk. - Term premium increases may lift banks’ funding costs and raise credit risk for households and firms. - Annex II econometric table: regressions for Austrian, CESEE, and Global groupings; Observations: 60 per specification; R-squared range: 0.087 to 0.813 across specifications. Selected significant coefficients preserved in source. - Annex V sovereign haircut calibration: - Shock derived from 90th percentile of historical annual changes of daily yields (Bloomberg generic 5-year yields 2005–12) for fifty countries; special treatments for Belarus, Bulgaria, Luxembourg, Malta, Romania, Cyprus, and Estonia. - Table reports large haircuts for Greece (e.g., 1Y: 125.29; 2Y: 168.69; other large entries), Portugal (e.g., 2Y: 16.15; 30Y: 87.76), and varied profiles across EU and CESEE countries. Selected country maturity haircuts preserved verbatim where shown. *Source: EXECUTIVE SUMMARY and chapters/extracts from IMF staff report (Austria FSAP Update) as provided in the supplied PDF excerpt.* ### EXECUTIVE SUMMARY ___________________________________________________________________________ 7 ### EXECUTIVE SUMMARY ### INTRODUCTION AND CONTEXT - The Austrian banking system is in a recovery phase following the 2008–2009 global financial crisis. - The 2008–2009 crisis led to full or partial nationalization of three mid-sized domestic banks and significant government liquidity and capital support. - The Financial Market Stability Act (FinStaG) authorized recapitalization up to €15 billion; as of Dec 2012, €13.6 billion were utilized, of which capital injections reached €5.9 billion. - The consolidated banking sector assets amounted to about €1.1 trillion, or more than 3.5 times GDP in 2012 Q3. - The three large internationally active banks account for almost half of total bank assets. - Bank business models are diverse: two broad groups each account for about 45 percent of banking system assets (large internationally active banks focused on CESEE, and domestic retail–focused banks); (partially) nationalized medium-sized banks represent about 7 percent of total assets. - Net interest income is the primary profit source; in 2012 net interest income amounted to almost three times the income from securities holdings. Return on assets stood at 0.3 percent in 2012. ### MACROECONOMIC OUTLOOK - Baseline forecasts for Austria and CESEE indicate a gradual macroeconomic upturn, but growth remains lower than pre-crisis levels. - CESEE medium-term growth prospects remain stronger than those for advanced economies; the CESEE baseline used is a BIS-weighted average of individual countries’ projections as of June 2012. - WEO projections used in the stress test are as of Oct 2012. ### SOLVENCY STRESS TESTS — MAIN RESULTS AND CAVEATS - Aggregate capital buffers are sufficient to withstand severe but plausible shocks from adverse macroeconomic developments. - Under the most severe scenario, the estimated total capital shortfall amounts to 1 percent of GDP. - The solvency stress test results reflect: - Comfortable initial capital buffers built in response to the crisis (de-risking of balance sheets and recapitalization via increased retained earnings). - Methodological limitations and asset-quality measurement challenges, especially in some CESEE countries; results should be interpreted with caution. - The three-year stress-testing horizon does not consider: - Repayment of state participation capital that benefited from a grandfathering clause under the Basel III phase-in transitional schedule (until 2018). - Potential implementation of a capital surcharge on domestic systemic institutions (from 2016 on). - The upcoming bank asset quality reviews by the ECB are expected to provide a more robust basis for assessing bank balance sheets and policy responses. - The banking sector appears well positioned to meet Basel III capital requirements: - On aggregate, the sector would comfortably pass the hurdle rates laid out by the Basel III phase-in arrangements for CET1 under the most severe scenario. - Capital buffers above the minimum Tier 1 capital ratio are somewhat thinner because Austrian banks hold limited amounts of non-common equity Tier 1 qualifying capital (private preferred stock and minority interests). ### LIQUIDITY STRESS TESTS — MAIN RESULTS - Austrian banks’ funding structure appears resilient across major currency buckets. - Under a severe 30-day funding stress scenario, the total liquidity shortfall is estimated at only 0.1 percent of total liabilities. - Foreign currency liquidity has substantially improved since 2008, though some banks need to continue efforts regarding CHF funding. - Improvements in liquidity position are attributed to enhanced liquidity supervision and monitoring by the OeNB and strengthened supervisory standards for banks’ liquidity risk management. ### CONTAGION AND FUNDING/NETWORK ANALYSIS - Network analysis indicates the Austrian banking system is robust to funding and contagion shocks. - Large banking groups do not experience losses due to: - Strong counterbalancing capacity. - The network structure of the Austrian interbank market. - The impact on capital adequacy for the whole banking system is not material and is driven primarily by fire sales rather than rising funding costs or contagion defaults. ### CROSS-BORDER SPILLOVERS AND SYSTEMIC RISK - Cross-border spillovers between Austrian banks and peer banks active in CESEE appear contained overall. - Severe distress affecting the top two Austrian banks can transmit to other banks in CESEE; on average this transmission is less than the systemic risk potentially introduced by severe distress affecting other CESEE peer banks. - Inward spillovers from CESEE peers to Austrian banks are not significantly different from those affecting other banks in the region. - Evidence suggests the need to combine micro-prudential and macro-prudential perspectives in regulation of systemic institutions due to a weak link between large European banks’ individual solvency risk and their estimated contribution to systemic risk. ### KEY RECOMMENDATIONS ON STRESS TESTING (HIGHLIGHTS FROM TABLE 1) - High priority, near-term: - Assess impact of different regulatory ratios across CESEE jurisdictions and potential ring-fencing of cross-border flows on consolidated regulatory ratios of Austrian banks. - Continue developing funding/contagion analysis by endogenizing fire sales and funding costs; allowing domino effects from cross-guarantee schemes and cross-holdings of unsecured paper; and extending bilateral exposure networks to global banks. - High priority, medium-term: - Consider modeling credit risk in CESEE based on insolvency data—subject to data availability—to avoid reliance on loan loss provisioning amid asset-quality concerns. - Medium priority, immediate to medium-term: - Include market risk factors, including rises in risk premia, in the satellite model for credit risk in domestic and cross-border exposures. - Develop a framework to identify and measure banks’ individual contribution to systemic risk drawing on market-based approaches such as CoVaR. - Contribute to approaches integrating solvency, liquidity, and market shocks in stress test scenario design, include second-round effects—particularly on credit growth—and allow for transmission of behavioral shocks. *Source: EXECUTIVE SUMMARY, Austria FSAP Update (IMF).* ### 6. Austrian banks have sizable cross-border linkages, especially in the CESEE region, but ### _cr1416 - 6. Austrian banks have sizable cross-border linkages, especially in the CESEE region, but ### Cross-border linkages and exposures - Direct and cross-border lending exposures amount to nearly €460 billion. - €326 billion of those exposures are to CESEE countries or under 30 percent of overall banking system assets. - Total assets of foreign subsidiaries in CESEE reached €234 billion, or over 70 percent of total exposure to the region, in Sep 2012. - This regional exposure is highly concentrated in the large internationally active banks (for over 80 percent of aggregate subsidiary assets). - Austrian banks are primary lenders in CESEE countries, with market shares above one-third in Slovakia, Bosnia, Romania, Albania, and the Czech Republic. - Foreign-owned banks in Austria represent more than 25 percent of the total banking system by assets, dominated by one large bank and two mid-sized banks. - Austrian banks’ exposure to European peripheral countries fell to €31 billion from €45 billion in 2008; of the €31 billion, 17 percent were claims on the public sector and 24 percent claims on credit institutions in these countries. ### Public debt, government support, and fiscal context - Austrian public debt is expected to peak in 2013 at around 74 percent of GDP. - This compares with the expected 109 percent ratio for advanced economies and 95 percent for the euro area. - Baseline projections show the public debt ratio will decline gradually towards pre-crisis levels of 60 percent of GDP supported by a medium-term reduction of general government debt. - Government support to the banking system has been significant, including through the nationalization of three medium-sized banks. - Uncertainties remain related to the restructuring of the nationalized banks and the realization of contingent liabilities. ### Liquidity developments and external support - Since 2008, banks have continuously improved their liquidity position across major funding currencies. - The increase in liquidity buffers has been mainly facilitated by the ECB’s monetary policy, the repo operations conducted by the SNB, and the swap facilities provided by the SNB and the ECB. - Austrian banks participated in the two ECB’s supplementary longer-term refinancing operations (Dec 2011, Feb 2012) with a total volume of EUR 15.7 billion, which corresponds to 1.5 percent of the total allotted volume (well below Austria’s proportionate Eurosystem share of 3.8 percent). - Reflecting a recent pick-up in deposit growth at Austrian banks, their dependence on ECB financing is, however, relative low relative to their euro zone peers. ### FSAP stress testing: objectives, coverage, and approach - Objective: assess resilience of the Austrian banking sector to adverse macroeconomic conditions and severe stress in global funding markets. - Solvency tests: - Top-down (TD) test undertaken by the OeNB collaboratively with the FSAP team on all 585 consolidated banks licensed in Austria. - Bottom-up (BU) solvency stress tests—focusing on market and sovereign risk—were run by the five largest banks (representing about 60 percent of banking system assets). - Liquidity stress test: - Covered the largest 29 banking institutions (accounting for around 80 percent of total assets). - Based on a range of adverse scenarios, broken down by major currency, with severe liquidity stress lasting for up to one year. - Contagion analysis: - A contagion module assessed potential for distress in an individual banking institution to create risks to overall financial stability. - Network model of the Austrian interbank market used to assess domino effects. - Contagion through financial markets evaluated using the CoVaR methodology. ### Key risk factors included in the stress tests - Deteriorating asset quality: sharp slowdown in Austria and CESEE could significantly worsen asset quality of domestic portfolios, cross-border operations, and subsidiaries’ loan books. - Declining profits: protracted CESEE slowdown could erode net interest margins; CESEE subsidiaries’ share in total consolidated net operating profits has fluctuated around 50 percent over 2006–2012. - Credit risk from foreign currency lending: exchange rate volatility (e.g., CHF) and repayment vehicle loans (RPVs) risks; high share of foreign-currency loans (FCLs) in CESEE may weigh on credit quality following domestic-currency sell-offs. - Sovereign risk: perceptions deteriorated during the financial crisis; lower valuations of CESEE government bonds and rising term premia could affect banks’ capital. - Market risk: widening credit spreads on European financial institutions or corporate entities could affect profitability via valuation effects or higher risk weights. - Securitization risk: rapid downgrades of structured credit products could impact capital adequacy ratios. - Funding/rollover risk: rising libor-ois spreads, dry-up of money markets, and FX swap market disruptions may increase refinancing costs or impair rollover of maturing contracts. - Financial contagion: fire sales, interbank market propagation, and contagion among similar business models may affect solvency/liquidity. - Regulatory changes: implementation of Basel III via CRD IV/CRR and the repayment of public participation capital (state aid) may add pressures; CRD IV/CRR expected to come into force in January 2014, full Basel III implementation scheduled for January 2019, phase-out of participation capital grandfathered until January 2018. ### Stress-test design, scenarios, and modeling specifics - Timeframe: solvency stress tests used supervisory and macroeconomic data as of end-2012 over 2013–2015; end-of-year supervisory consolidated data became available in May 2013. - Two-pronged solvency approach: - Top-down tests: all 585 banks, based on quarterly baseline projections from the Austrian Quarterly Model (AQM). - Light bottom-up tests: five largest banks, focusing on market and sovereign risk. - Country coverage and granularity: - Country-specific macro scenarios generated for twenty-two CESEE countries using OeNB G-VAR model covering 51 countries and the euro area (estimated over 1995–2012). - Adverse scenarios applied a double-dip shock to real GDP growth from baseline over the first two years with positive adjustment in the last year. - Adverse scenarios considered: - Global shock and intensification of the euro area economic crisis: two-standard deviation shock to Austrian GDP growth and spillovers to CESEE/CIS leading to a deviation from baseline growth of one and a half standard deviation across the region. - Severe recession in CESEE/CIS: aggravated downturns bringing trend regional growth down by 1.8 standard deviations together with a two-standard deviation shock to Austrian GDP growth. - Global funding scenario: reflects acute stress conditions observed in late 2008, including increased funding costs and restricted FX swap market access. - Macro projection example: - For the Austrian economy, the 2-year cumulative growth over 2013-2014 is projected at 3.05 percent under the baseline scenario and at -2.97 percent under the adverse scenario. ### Modeling and market-based analysis - Complementary econometric analysis on credit risk of the largest listed banks used monthly changes in CDS spreads and Moody’s KMV expected default frequencies (EDF) from October 2007 to October 2012. - Risk-factor categories in regression analysis: - Austria macro-financial variables. - Contagion from CESEE sub-regions (NMS-04, NMS-07, SEE, CIS). - Global risk factors, including time-varying risk premia from US equity and fixed income markets (equity premium, volatility risk premium, term premium). - Network, CoVaR, and other modules used to capture contagion via interbank links and market channels. *Source: IMF technical note (FSAP chapter) as provided in the content unit.* ### 20. The approach to credit risk modeling as a function of macroeconomic developments ### 20. The approach to credit risk modeling as a function of macroeconomic developments ### Credit risk modeling framework (domestic vs cross-border) - Exposure at default (EAD) from domestic and cross-border credit stood at 56 percent and 44 percent, respectively. - Exposures to the CESEE region accounted for about 70 percent of all cross-border exposures. - For domestic exposures: - A credit risk model links sectoral corporate probabilities of default (PD) in six Austrian corporate sectors to a wide range of observable macroeconomic variables and a latent risk factor (Box 1). - For cross-border operations and foreign subsidiaries: - A separate satellite model based on country specific CESEE loan loss provisioning ratios (stock and flow ratios) is calibrated to assess credit risk in cross-border operations and foreign subsidiaries. - Changes in provisioning ratios are used to proxy changes in PDs. ### Stressed Loss Given Default (LGD) estimation - Stressed LGD is estimated separately for collateralized and uncollateralized exposures. - Real estate collateral: - Country-specific haircuts are estimated for CESEE countries based on the elasticity of GDP growth to house prices and the GDP growth path projections under each scenario. - Uncollateralized exposures: - A country-specific LGD, capped at 45 percent for Austria, and linked to the World Bank Doing Business Statistics for CESEE countries—with the distribution truncated at 80 percent—is generated under the baseline scenario. - This value is stressed through linear increments each quarter reaching a final add-on of 10 percentage points in 2015Q4 under the adverse scenario. ### Profit dynamics, FX depreciation, and balance sheet contraction - During the stress test horizon profits decline significantly mainly driven by a depreciation of host country currencies and the contraction of banks’ balance sheet. - Scenario calibration and impacts: - Weak macroeconomic performance triggers depreciation of local CESEE currencies relative to the Euro. - The sustained cumulative depreciation reaches 1.5 (2.0) SD under the adverse (severe) scenario in 2013–2014, with a partial and gradual rebound assumed during the last year of the stress test horizon. - CESEE subsidiaries' share in total consolidated net operating profits reached over 50 percent in 2012 Q2. - CESEE subsidiaries' share in total consolidated assets stood below 25 percent in 2012Q2. - Operating profits decline further triggered by a drop in net interest income caused by performing loans becoming non-performing. - It is assumed that all the components of operating profits decline in line with net interest income. ### OeNB’s Credit Risk Model for the Austrian Economy (Box 1) - Endogenous variables: quarterly default frequency rates over 1985-2011. - Corporate sectors: construction, production, trade, transport, tourism and services. - Explanatory variables: nineteen macroeconomic time series; up to six quarterly lags per variable. - Model selection: - Forward-Stepwise Selection algorithm applied; for each number of regressors the best five models are selected by explained sum of squares. - Each model estimated with an unobserved component reflecting a latent risk factor. - Key equation components preserved as presented: logit-transformed sectoral default frequency, k macroeconomic variables x_j, unobserved factor z_i, error terms ε_i and v_i. - Findings: - Aggregate credit risk is driven by common variables (e.g., inflation, interest rates, credit growth) and sector-specific variables (e.g., exports, capital investment, oil prices). - Latent risk factor is significant only in small models; for models with more than seven macroeconomic variables, evidence for a latent risk factor vanishes. - Conclusion: given a broad macroeconomic dataset, Austrian models do not have to rely on latent factors. ### Sensitivity analysis — foreign currency lending (FCLs) - Foreign currency loans (FCLs) pose additional risk due to declining ability to pay of unhedged borrowers following appreciation of foreign currency. - Analysis conducted separately for domestic and CESEE exposures; approach depends on loan structure and data availability. - Domestic FCL stock and composition (as of 2012 Q3): - FCLs to domestic non-banks amounted to €50.7 billion, corresponding to 15.3 percent of all domestic loans. - Of these, €34.6 billion were owned by households (share of 25 percent of housing loans) and €10.0 billion by corporates (share of 7 percent of loans to non-financial corporates). - Loan structure: - About 60 percent of FCLs to households and corporates are arranged as bullet loans associated with repayment vehicles (RPVs), with 40 percent being amortizing loans. - From the €30.0 billion bullet loans outstanding in September 2012, €25.8 billion were FCLs, of which €24.0 billion were denominated in CHF. - 60 percent of all FCLs are linked to RPVs. - As of 2012Q3, almost 80 percent of FCLs associated to a RPV showed a maturity beyond 2020. - Provisioning: - Provisioning rate on FCLs granted to non-banks stood at 1.1 percent in 2012 Q3, less than one third of that associated to euro loans. ### Swiss Franc FX sensitivity scenario - The analysis assumes a protracted appreciation of the Swiss Franc vis-à-vis the Euro for Swiss Franc loans (about 90 percent of FCLs to households and corporates are denominated in Swiss Francs). - Assumption: Swiss Franc appreciates by 1.5 SD over 2012Q3–2014Q4 with the nominal exchange rate climbing from 1.21 to 1.07 and stabilizing in the last year of the stress test horizon. - This is equivalent to an appreciation of the Swiss Franc vis-à-vis the Euro of 13 percent over 2012Q3–2014Q4. - RPVs and market risk: - RPVs are exposed to market risks; about three quarters of RPVs are directly linked to capital market developments with over half of outstanding loans linked to mutual funds-based life insurance instruments. - A separate sensitivity test examined the potential rise in estimated projections of baseline funding gaps of RPVs under adverse market developments; results interpreted with caution due to extensive assumptions and data unknowns. ### Indirect credit risk modeling from FCLs (Box 2) - Assumes FCL borrowers are unhedged. - Appreciation of CHF increases value of outstanding debt in EUR by FXDΔ* (FX = nominal exchange rate of EUR per CHF), increasing debt-to-income ratio by (FXDΔ*)/I which affects credit losses with an elasticity estimated at 2.5. - Relationship used: - ΔLLPR = −5.2 * (ΔD* / I) [presented as FX I D LLPR Δ − = Δ **5.2 in source] - Exchange rate path assumption: - Protracted appreciation of the Swiss Franc vis-à-vis the Euro with volatility equal to 1.5 SD. - Exchange rate path driven by: FX_t = FX_0 * exp*(1.5*σ*sqrt(t)) [presented in source as σ*5.1*0 exp* t t FXFX =] - Additional impairments from the equation are distributed equally over each loan’s remaining maturity. - Stress test loss is the accumulated loss over the stress test horizon 2012Q4-2013Q1. ### CESEE FCL sensitivity approach - Credit risk from FCLs in CESEE assessed using impairment data broken down by currency. - An “FX boost factor” defined as the elasticity of changes in loan loss provisioning rates (LLPRs) relative to local currency loans triggered by an appreciation of the FX is applied to all FCLs in Swiss Francs. - The elasticity is the average estimate of fifteen different models fitting excess LLPRs of FCLs to changes in the FX, assuming non-linear functional forms (exponential and quadratic) and different selection criteria. ### Securitization risk - Stress test on securitization positions applied via an increase in risk weighted assets (RWAs). - Due to opacity and non-linear payoffs, a credit risk migration matrix is assumed in line with the baseline scenario calibration of the 2011 EBA stress test exercise. - Migration matrices calculated separately for medium-risk and high-risk positions. - Stressed risk weights computed as a weighted average of original risk weights and migration factors. - Impact on capital ratios: - Defaulted exposures are 1,250 percent risk-weighted. - Stressed risk weights from securitization positions are combined with those from non-securitization assets to compute RWAs. ### Sovereign risk - Measured in the adverse scenario through changes in sovereign yields leading to repricing of all affected bonds (both banking book and trading book). - Scope of sovereign includes: all central governments (but no central banks), all regional governments, and all local authorities. - Calibration and application: - Term structure of sovereign risk shifts upwards for all countries to which Austrian banks are exposed, including sovereign bonds held by CESEE subsidiaries to comply with local liquidity requirements. - Haircuts to the banking book are applied to adjusted (marked-to-market) balance sheet values. - Approach allows for changes in term premia (slope and curvature) rather than a parallel shift on spreads. - Calibration for fifty eight countries: - For fifty countries, shock derived from 90th percentile of historical distribution of annual changes of daily yields (3-month to 30-year) using Bloomberg-based generic 5-year government bond yields over 2005–12. - For Belarus, Bulgaria, Luxembourg, Malta, and Romania, haircut computed using extreme returns for the most liquid outstanding international bond as of Dec 2012. - For Cyprus, haircut calibrated from the sovereign yield curve as of Dec 2012. - For Estonia only international loans were outstanding at end-2012. ### Market risk - Market risk sensitivity analyses run by the largest five banks as part of the “light bottom-up approach” using trading book positions as of 31 December 2012. - Valuation effects reported for each risk category individually and in total. - Calibration consistency with macro stress scenario: - Volatilities of daily changes in interest rates were scaled by a factor of two and were additionally scaled over a one-year time horizon. - Eastern Europe interest rates determined by an equally weighted basket of CZK and PLN. - Other non-emerging markets’ interest rates determined by an equally weighted basket of CHF, JPY and AUD. - FX rates defined vis-à-vis the Euro with scenario assuming a Euro depreciation in line with the solvency stress test and FCL sensitivity analysis. - Emerging Markets FX defined as an equally weighted basket of eight CESEE/CIS countries. - Other non-emerging markets’ FX determined by an equally weighted basket of CHF, AUD and CAD. - FX volatilities follow EBA 2011 calibration for rates involving major currencies and for rates involving at least one Emerging Market currency. *Source: _cr1416 - 20. The approach to credit risk modeling as a function of macroeconomic developments* ### 39. Other risk categories include the credit risk factors identified in the FSAP team ### _cr1416 - 39. Other risk categories include the credit risk factors identified in the FSAP team ### Credit risk and market risk factors - Credit risk factors included in sensitivity analysis: itraxx high yield Europe, itraxx Europe crossover, itraxx senior financials. - Market risk parameters calibrated in line with the adverse EBA 2011 stress test scenario (Equity, Funds, Commodities, Counterparty). ### Basel III Implementation - Full implementation of Basel III requirements, including front-loading of phase-in capital arrangements and Basel III RWAs, estimated impacts on projected regulatory ratios: - CET1 would shed 1.4 percentage points relative to EBA CT1 capital for the whole banking system and 1.6 percentage points for the large international banks. - Impact on Basel III Tier 1: 1.7 percentage points for the system and 2.2 percentage points for the large international banks. - Impact on the total adequacy ratio (CAR) depends on treatment of €19.8 billion (€10.0 billion) long-term subordinated debt held by the banking system (large international banks); effect on the CAR could range between 1.9 and 4.0 (2.3 and 4.9) percentage points for the banking system (large international banks). - Main driver of the decline in CET1: phase-out of participation capital subscribed by the government in the context of state aid (€4.1 billion for the whole banking sector of which €3.0 billion was issued by the largest international banks), kicking in January 2018. ### Solvency Stress Test Results - Overall conclusion: Austrian banks benefit from sufficient capital buffers, including under most adverse circumstances. - Key findings under the most severe macroeconomic scenario: - Banks representing less than 7 percent of total bank assets would fall below the regulatory threshold. - Estimated aggregate capital needed to restore these banks above regulatory minimum: 0.3 percent of total bank assets, or about 1 percent of GDP. - Percent of total bank assets in the 6–8 percent core Tier 1 bucket increases from about 4.5 percent under the baseline to 17 percent under the severe scenario at end of stress horizon. - Projected losses under the severe scenario dent large banks’ core Tier 1 ratio by 4.8 percentage points versus average 2.8 percentage point losses for the Austrian banking system. - Drivers of regulatory capital decline: - Credit risk in CESEE is the main driver. - Projected profit and loss effect from severe macroeconomic stress adds up to 1.5 percentage points driven by credit risk effects. - More than two-thirds of expected losses from credit risk come from CESEE exposures; under one-fourth originate domestically. - Full Basel III implementation (including front-loading) would have an impact of 1.4 percentage points led by phase-out of eligible capital. - Large internationally-active Austrian banks: - Remain on aggregate above the regulatory hurdle even under full implementation of Basel III qualitative phase-in arrangements. - Top three banks continued building capital buffers to be compliant with full implementation of qualitative and quantitative Basel III rules for CET1 as of 2012 Q4. - Projected capital ratios for large international banks are above transitional quantitative hurdle rates at end-2015, even under the most severe scenario. - Additional sensitivity analysis (not related to main scenarios): - Indirect credit risk from FX appreciation in foreign currency loans (Austria and CESEE/CIS) adds up to 35 bps in CT1 ratios for the whole banking sector. - Market risk from underperformance of market instruments associated to RPVs in domestic bullet loans contributes an additional 7 bps. - Estimated impact from securitization exposure risk: 10 bps. - Business-unit (BU) stress tests for largest five banks: - Credit risk from FCL has greater impact on capital adequacy than combined sovereign and market risk. - Additional credit risk from FCL would depress CT1 capital ratios for the five banks by 41 bps under the most adverse scenario. - Sovereign risk has additional impact of 51 bps (estimates are conservative as portfolio correlations across high-yield assets and safe haven bonds were ignored, potentially overestimating sovereign risk). - Market losses from adverse shocks across interest rates, FX, equity, commodities, high yield credit risk, and CVA are insignificant. ### Liquidity Stress Tests - Scope and methodology: - Conducted on a consolidated basis for the largest twenty nine banks (accounting for over 80 percent of total banking system assets), using 2012Q4 data and forward-looking shocks on contractual and behavioral cash flows across five maturity buckets (up to 5 days, 1m, 3m, 6m, 12m). - Stressed haircuts calculated on reported collateral value after central bank haircut; haircuts vary by central bank due to eligibility differences. - Tests conducted separately for six currency buckets: EUR, USD, CHF, GBP, JPY, other. - Time horizons, scenarios, and counterbalancing capacity approaches: - Three horizons: 30-day, 90-day, one-year. - Five market scenarios: baseline, market mild, market medium, market severe, combined (includes idiosyncratic shock on rollover of retail and wholesale deposits plus capital market shock). - Three counterbalancing capacity severity approaches: - Full counterbalancing capacity: includes less liquid assets evaluated at baseline haircuts but excludes committed lines and parent injections. - Increased focus on market liquidity: haircuts for less liquid assets increase to 100 percent. - Market liquidity: haircuts on unencumbered eligible assets at the Eurosystem increase to 100 percent for securities with ratings below A- (in addition to ‘Increased focus on market liquidity’). - Combined scenario behavioral assumptions: - 30-day horizon: expected rollover rates drop to 90 percent (wholesale) and 95 percent (retail); banks cut expected financial investments by 50 percent; do not cut expected new loans. - 90-day horizon: rollover rates decline to 80 percent (wholesale) and 90 percent (retail); banks shed new investments by 100 percent. - One-year horizon: drawings of committed lines by households and non-financial corporates jump by 100 percent; non-financial loans reduced by 4 percent; new unsecured loans to other banks drop by 100 percent. - Calibration details: - For collateral deposited at the Eurosystem, haircuts reach up to 100 percent under the market severe scenario and 30-day horizon depending on issuer rating and asset class. - For foreign currency collateral, haircuts under market severe scenario are 10 percent (30-day) and 15 percent (90-day). - For tradable assets not deposited with central banks, additional haircuts are a function of security rating (applied haircuts include: 5 percent (AAA), 10 percent (AA), 15 percent (A), 50 percent (BBB), 70 percent (other pledgable assets), 100 percent (committed lines and parent liquidity injections)). - PD shifts: satellite model yields increase of 110 percent PDs under adverse scenario on an annual basis implying 11 percent shift over 30 days; haircuts on non-tradable collateral linked to estimated PD shifts and increase over the liquidity stress horizon. - Market reaction and issuance: - First quarter: liquidity dries up for all issuers irrespective of collateral quality; investors cannot differentiate across issuers/instruments. - Second quarter: banks with core Tier I ratios of at least 10 percent under the most severe scenario or that increased core Tier 1 by more than one percentage point regain partial access to funding markets and can place 50 percent of expected issuance. - Comparison with Basel III LCR: - FSAP liquidity stress test is more stringent than LCR: extends to one-year horizon; requires currency- and maturity-specific inflows; incorporates stricter forward-looking assumptions though with less severe wholesale deposit run-off in some LCR revisions; stressed haircuts more severe (unencumbered eligible assets at Eurosystem increase to up to 100 percent for securities below A-). - Aggregate liquidity shortfall results under severe assumptions (closure of unsecured interbank and FX swap markets and substantial haircuts): - Total liquidity shortfall based on cumulated counterbalancing capacity amounts to 0.1 percent (30 day horizon), 0.3 percent (90 day horizon), and 0.2 percent (1 year scenario) of total liabilities of the twenty nine banks in the sample. - Foreign currency liquidity: - Substantial improvement in USD and CHF liquidity positions since 2008; resilience to USD and CHF funding shocks is now high. - CHF funding: half of banks made substantial progress; the other half need to diversify funding sources and lengthen funding tenors given CHF asset maturity structure and limits to reducing the portfolio. ### Contagion Analysis - Two initial conditioning events modeled: - Rising funding pressures: ‘global funding scenario’ replicating post-Lehman liquidity strains (sharp rise in funding cost, credit freezes). Banks under liquidity squeeze engage in fire sales, eroding capital buffers; post-shock capital combined with network model simulates cascading defaults in interbank market. - Drop in market value: repricing of market risk factors causes portfolio/credit losses pushing a bank’s returns to the left tail of the distribution; bank reaches its VaR returns in the market-implied value of assets. *Source: OeNB and IMF staff estimates as presented in the FSAP chapter.* ### 56. The transmission of each separate initial shock from an individual bank to the broader ### _cr1416 - 56. The transmission of each separate initial shock from an individual bank to the broader ### Transmission channels - Bilateral Exposure: - Counterparties with a significant exposure to the failing firm may suffer material losses resulting in their inability to satisfy their obligations thus transmitting distress to other parts of the financial system down the credit chain in the form of cascading defaults. - Market Contagion: - Market participants’ revise their expectations on the solvency of other firms following similar business models than the firm in distress, conditional on the broader economic environment. ### A. The Funding/Network Analysis — methodology and scenario - Purpose: - Capture solvency effects from negative liquidity gaps and potential default cascades triggered by an insolvent firm on its creditors. - Data and model: - Funding/network analysis draws on Austrian banks’ bilateral matrix of exposures. - Domino effects estimated using a default cascade model for the consolidated Austrian interbank market (Furfine’s network model). - Bilateral netting of intra-group exposures is allowed. - Assumed LGD on net exposures is 100 percent. - A bank is considered in default when its capital adequacy ratio at the consolidated level falls below 8 percent. - One-year global funding scenario (mimics post-Lehman liquidity shock): - New issuance: unsecured interbank, FX swap markets and capital markets close. - For the very strong banks, markets reopen gradually allowing up to 70 percent of banks’ expected issuance after the first quarter of liquidity stress. - Net cash-outflows: - Contingent liabilities rise triggered by a 50 percent jump in the drawings of committed lines. - Inflows from loans/maturing paper in banks’ own portfolio decrease in line with the PD shifts estimated under the solvency test. - Counterbalancing capacity: - Liquid assets suffer from sharp price declines driving up haircuts across asset classes. - For assets not deposited at the central bank, granular haircuts range between 12.25 percent (investment grade) to 80 percent (sub-investment grade). - Cost of funding increases: - Cost of expiring term retail funding increases by about 140 bps. - For those few banks that are able to access capital markets, the cost of new issuance rises by 70 bps. - Behavioral assumptions: - Banks cut their expected financial investments by 40 percent. - Banks’ expected new unsecured loans to other banks are reduced by 100 percent. - Repos on general collateral are assumed to be rolled over. - The analysis assumes that banks cannot draw committed liquidity lines nor receive liquidity injections from their parent bank. - Fire sales and contagion mechanics: - Banks with negative funding gaps engage in distress sales; stressed value of assets computed using counterbalancing capacity haircuts. - Banks that become illiquid incur asset fire losses up to the depletion of their counterbalancing capacity. - Contagion mechanism is based on a network approach of interbank exposures; creditor losses realized immediately with recovery rates of 0 percent. ### A. The Funding/Network Analysis — key findings - System resilience: - The Austrian banking system is resilient to potential funding and contagion stress transmitted through creditors’ exposures in the Austrian interbank market. - Under the global funding scenario, large banking groups do not experience losses due to their strong counterbalancing capacity and the network structure of the Austrian interbank market. - Aggregate impact: - The impact on the core Tier I ratio of the whole banking system is limited to 108 bps, and is driven primarily by banks’ fire sales rather than by cost of funding effects or cascading defaults. - Caveats to interpretation: - Fire-sale assets are calibrated exogenously; spiral effects from further declines in prices as a function of the aggregate increase in supply of assets are not modeled explicitly. - Mark-to-market effects from common exposures to stressed assets by banks holding similar assets are not computed. - Domino effects outside the Austrian interbank market (e.g., inability to service other debt instruments or stress of counterparties domiciled outside Austria) are not considered. - Contagion effects from a bear-market sentiment to banks following similar business models to the bank in distress are excluded. ### B. The CoVaR Analysis — methodology and scope - Purpose: - Complement the network analysis by capturing contagion through market channels and common leverage/comovement not captured by bilateral credit exposures. - Features: - Uses combined market and balance sheet-based approach; contagion effects from Austrian banks’ left tail comovement in leverage ratios and financial returns assessed using the CoVaR methodology. - Distress is defined as a firm reaching its VaR. - Two peer groups constructed: - A European banking system including forty internationally active banks. - A CESEE banking system formed by fourteen large foreign banks active in the CESEE region. - Analysis period: weekly market valued asset returns over April 2005–Dec 2012. - Time-varying VaR/CoVaR dynamics characterized using European and US financial risk factors as state variables. - Channel emphasized: - Contagion through financial markets and changes in banks’ leverage even without significant creditors’ exposure to the distressed firm. - Global banks can transmit instability to/from large international Austrian banks acting in the CESEE region or in global funding markets. ### B. The CoVaR Analysis — key findings - Predictive state variables: - Among financial variables, liquidity strains exhibit the strongest predictive power to forecast European and CESEE banking system tail returns. - European banks’ left tail returns are also affected by an uptick in equity market volatility, changes in the short-end of the yield curve, and the sovereign debt crisis. - Asymmetry of spillovers: - Effect of individual banks’ distress on banking system tail returns is very significant and asymmetric. - For the European banking system, the coefficient of individual negative returns on banking system returns is more than four times larger than the coefficient of positive returns. - For the CESEE peer group, it is over two times larger. - Cross-border spillovers: - Outward spillovers from top two Austrian banks to other banks in the CESEE region are not negligible but are less than the systemic risk potentially introduced by severe distress affecting other CESEE peer banks. - Inward spillovers to Austrian banks are most pronounced from banks headquartered in the CESEE region that are active across CESEE or foreign banks with the highest presence in the region. - Severity of distress transmitted to Austrian banks by CESEE peers does not appear significantly different than for other banks. - Micro vs. macro prudential implications: - Weak link between individual and systemic risk: - No correlation between banks’ risk in isolation (VaR) and banks’ contribution to systemic risk (∆CoVaR) for both European and CESEE systems. - Policy implication: - Any add-on on regulatory capital requirements designed to contain spillover effects needs calibration drawing on both individual and systemic risk analysis. ### Conclusions and recommendations — summary findings and policy implications - Capital adequacy and exposure: - Stress test results suggest that the Austrian banking system is adequately capitalized against adverse macroeconomic shocks. - Under the most severe macroeconomic scenario: - Banks representing less than 7 percent of total bank assets would fall below the regulatory threshold. - Estimated aggregate capital needed to bring back the capital ratios of these banks above the regulatory minimum amounts to 0.3 percent of total bank assets, or about 1 percent of GDP. - Aggregate results mask heterogeneity; some mid-and small-sized banks are relatively more vulnerable. - Basel III positioning: - Banking sector appears well positioned to meet Basel III capital requirements. - Front-loading full implementation of Basel III (including total phase-in of deductions from CET1, enhanced risk-weighted assets, and a 2.5 percent capital conservation buffer) would have an impact of 1.4 percentage points on core Tier I capital, on top of 1.7 percentage point impact from projected profit and loss effects from severe macroeconomic stress. - Capital buffers above minimum Tier 1 are thinner due to limited non-common equity Tier 1 qualifying capital (private participation capital and minority interests). - Cautions and limitations: - Asset quality—particularly in some CESEE countries—is still deteriorating and difficult to assess with full confidence; upcoming bank asset quality reviews by the ECB should provide a more robust basis. - The three-year stress testing horizon does not consider the repayment of state participation capital. - Potential for a capital surcharge of up to 2.0 percent of RWAs laid out by forthcoming CRD IV/CRR on domestic systemic institutions lies outside the scope of the stress testing exercise. - Stress tests use Basel III regulatory minimum capital requirements as hurdle rates which may lie below market-expected capital buffers to keep funding costs low. - Other caveats include reliance on reduced-form approaches and estimated quantitative relationships that may not hold in periods of extreme stress. - Stress tests are based on consolidated supervisory data and do not allow for capital or liquidity ring-fencing measures in host countries which could constrain free allocation of capital and funding relationships between foreign subsidiaries and parent Austrian banks. *Source: IMF staff report content provided in the supplied PDF excerpt.* ### 78. Banks should continue building capital buffers gradually to support market ### _cr1416 - 78. Banks should continue building capital buffers gradually to support market ### Capital buffers and market confidence - Banks should continue building capital buffers gradually to support market confidence. - Risks that could undermine confidence include: - A confluence of adverse events. - Market expected buffers well beyond regulatory ratios. - Implementation of restrictions on cross-border transfer of capital across jurisdictions. - Additional capital requirements may be imposed on Austrian banks by investors to ensure pace with peers and to account for: - Their specific business model and comparatively large exposures to the CESEE region, where underlying asset quality is subject to significant uncertainty. - Potential exacerbating factors: - Increase in financial market fragmentation across the region. - Introduction of capital or liquidity ring-fencing measures in host countries that would constrain the free allocation of capital and funding relationships between foreign subsidiaries and parent Austrian banks. - Relevant regulatory and review timelines and provisions mentioned: - Basel III will be fully implemented in January 2019. - Before taking full responsibility for supervising euro-zone banks in the autumn of 2014, the ECB will undertake an asset-quality review, which will scrutinize the balance-sheets of 130 big banks, including the eight largest Austrian banks. - The phase-out of participation capital (private and public) subscribed in the context of state aid has been grandfathered under the Basel III phase-in transitional schedule until January 2018. - The CRD IV/CRR, expected to come into force in January 2014, empowers national authorities to set the so-called ‘Other-Systemic Institution’ capital surcharge of up to 2 percent from 2016 onwards. ### Liquidity and funding resilience - Banks’ funding structure appears resilient to the full implementation of Basel III liquidity coverage ratio. - Improvements attributed to: - Enhanced liquidity supervision and monitoring conducted by the OeNB. - Strengthened supervisory standards of banks’ liquidity risk management. - Local liquidity supervision requirements are significantly more stringent than the revised liquidity risk framework of Basel III and are applied across currency buckets. - Recommendation: - Banks should continue lengthening funding tenors in foreign exchange to be well prepared to withstand any funding shock in global markets. ### Stress tests, contagion, and systemic calibration - Network analysis findings: - Under a global funding scenario, large banking groups do not experience losses due to strong counterbalancing capacity and the network structure of the Austrian interbank market. - The impact on the CT1 ratio of the whole banking system is limited and driven primarily by fire sales rather than cost of funding or contagion effects. - Risks of cross-border spillovers between Austrian and other peer banks active in the CESEE region appear limited. - Policy use: - The contagion analysis can serve as an input in the calibration of domestic capital surcharges for domestic systemic banks under the CRD IV/CRR regulatory framework. - The network analysis informs how distress in a particular financial institution can be transmitted throughout the Austrian interbank market. - Contagion through market signaling or wrong-way risk can be measured using a market-based approach such as CoVaR. - Both network analysis and market-based tools can serve to monitor systemic risk and to identify rapidly growing activities that may pose systemic risk for early intervention. ### Stress test outputs and metrics (selected numeric items and scenarios) - Asset-quality review scope: 130 big banks, including the eight largest Austrian banks. - Regulatory dates and limits: - Basel III fully implemented in January 2019. - Phase-out of participation capital grandfathered until January 2018. - CRD IV/CRR expected to come into force in January 2014; ‘Other-Systemic Institution’ surcharge up to 2 percent from 2016. - Stress test scenarios referenced: - Baseline Scenario. - Global Shock Scenario. - Global Shock/Recession in CESEE Scenario. - Reported ratio ranges and chart references (as presented in figures): - Core Tier I, Tier I, and Total Capital ratio distributions reported across 2012Q4–2015Q4 with percentile box plots and hurdle lines. - CT1, Tier I, and Total Capital ratio buckets include categories: > 14%, 12-14%, 10-12%, 8-10%, 6-8%, 4-6%, 2-4%, 0-2%, < 0%. - Weighted-average Core Tier I, Tier I, and Total Capital ratios presented for 2012–2015 with value axes showing: - Core Tier I: 5.0% to 13.0%. - Tier I: 6.0% to 13.0%. - Total Capital: 8.0% to 15.0%. ### Market and liquidity stress-test parameter highlights - Market risk parameters (selected entries preserved verbatim): - Interest Rates — Advanced Countries: - USD 3Mbp10 - USD 2Ybp50 - USD 10Ybp150 - EUR 3Mbp30 - EUR 2Ybp70 - EUR 10Ybp130 - UK 3Mbp20 - UK 2Ybp90 - UK 10Ybp170 - Other non-Emerging Markets 3Mbp60 - Other non-Emerging Markets 2Ybp100 - Other non-Emerging Markets 10Ybp110 - Emerging Markets: - Eastern Europe 3Mbp100 - Eastern Europe 2Ybp110 - Eastern Europe 10Ybp140 - Asia 3Mbp40 - Asia 2Ybp40 - Asia 10Ybp50 - Volatilities: - US Volatility%40 - EUR Volatility%40 - UK Volatility%40 - Other non-Emerging Markets Volatility%70 - Emerging Markets Volatility%70 - Foreign Exchange FX Rates: - USD/EUR%18 - JPY/EUR%25 - GBP/EUR%12 - Other non-Emerging Markets currencies/EU%17 - Emerging Markets currencies/EUR%15 - Selected asset shocks: - Gold/USD%-20 - Equity Europe (Eursostoxx 50)%-15 - US (S&P 500)%-10 - Japan (Nikkei)%-20 - Ems (MSCI)%-20 - Liquidity Stress Test Medium Scenario Parameters (selected calibrations preserved verbatim): - Cash-Inflows: - Complete dry-up of unsecured interbank market (100%). - (Complete) dry-up of FX-swap market (30/90 days: 100%; 1 year: 20%). - Issuance of long-term unsecured bonds (30/90 days: -50%; 1 year: -20%). - Issuance of short-term unsecured paper (30/90 days: -50%; 1 year: -20%). - Issuance of long-term covered bonds (30/90 days: -50%; 1 year: -20%). - Issuance of short-term secured paper (30/90 days: -50%; 1 year: -20%). - Reduction of inflows from maturing loans & from paper in own portfolio maturing (based on a link to the solvency stress test). - Cash-Outflows: - 30/90 day horizons: Wholesale and retail deposits (bank individual values based on conservative assessment, average across banks -2%). - An additional idyosincratic shock reduces expected rollover rates of wholesale deposits to 90% and retail deposits to 95% over the 30 day period. - 1 year horizon: wholesale and retail outflows calibrated on peripheral Euro area experience during sovereign debt crisis (wholesale -3%, retail -2%). - Draw down of committed lines by other banks and non-banks (+50%). - Counter-balancing capacity: - Haircuts on non-marketable unencumbered collateral deposited at OeNB increase to 100%. - Haircuts on collateral deposited at OeNB with rating below A- increase to 100%. - Haircuts on unsecured issuances by banks and financial corporates as well as Asset Backed Securities deposited at the OeNB increase to 100%. - Haircuts on unencumbered collateral in USD, CHF, JPY, GBP, Other Currencies deposited at central banks (after CB haircuts) increase by additional 500 bp. - Haircuts on marketable collateral AAA to A- (not pledged at central banks) increase by 100 bp, 500 bp, and 1000 bp, respectively. - Haircuts on committed liquidity lines received by respective banks and liquidity injections from parent bank increase to 100%. - Banks behavioural reactions are factored into the 1 year scenario (reduction of unsecured interbank loans granted -100%, reinvestment of maturing liquid assets -50%, new loans granted -3%). - Haircuts on unencumbered marketable collateral with rating from AAA to A- pledged at OeNB increase by between 100 bp (asset category 1, credit quality step 1) and 1500 bp (asset category 3 and credit quality step 2) on top of OeNB standard haircuts; other asset categories and credit quality step collateral receive 100% haircuts. - Haircuts on BBB rated (or below) non-financial corporate bonds, other marketable, pledgable (but not pledged assets) collateral, and other non-marketable pledgeable collateral (but not pledged) increase to 100%. ### Risk Assessment Matrix (selected entries) - 1. Global shock/Intensification of the Euro Area debt crisis - Overall Level of Concern: Medium - Likelihood of Realization of Threat in the Next 1–3 Years: The IMF's Global Risk Assessment Matrix assigns a low/medium probability to a sharp slowdown in global growth or subdued domestic demand in the hard-hit euro zone periphery countries. - Expected Impact on Financial Stability if Threat is Realized: High - Key impacts described: - Decrease in demand for Austrian exports and trigger slowdown in the CESEE/CIS region, resulting in higher NPLs, lower profitability, and potential solvency pressures in some institutions. - Fiscal impact of financial sector-related public liabilities may trigger adverse market dynamics. - Tight funding conditions could affect some Austrian banks if European banks accelerate deleveraging, hoard liquidity and cut interbank lending. - Safe haven strategies may lead to sudden CHF appreciation, heightening credit risk from FX lending to domestic borrowers. - Authorities actions noted: Steps to promote local funding and to decrease the flow of FX lending in Austria. - 2. Contagion to/from CESEE countries - Overall Level of Concern: Medium - Likelihood of Realization of Threat in the Next 1–3 Years: Medium - Expected Impact on Financial Stability if Threat is Realized: High - Key points: - Highest Austrian bank exposures are to the Czech Republic, Croatia, Romania, Slovakia and Hungary. - Austrian banking system is a major gateway to CESEE via significant cross-border lending. - A sharp slowdown in CESEE would likely result in higher NPLs, lower profitability, potential solvency problems, and local currency depreciation in host countries could raise debt burdens of borrowers. - 3. Severe Funding Stress of Global Banks - Overall Level of Concern: Medium - Likelihood of Realization of Threat in the Next 1–3 Years: Medium - Expected Impact on Financial Stability if Threat is Realized: Medium - Key points: - Risk of renewed disruption in international funding and foreign currency swap markets. - Significant net cash outflows would trigger higher haircuts and unexpected margin calls on Austrian banks' securities portfolios. - Effect particularly severe for banks with negative funding gaps in USD and CHF. - Liquidity support from the ECB, repo operations and swap facilities by the SNB and the ECB may mitigate funding pressures. - FX-liquidity position of the system has substantially improved since 2008 through the lengthening of tenors and diversification of counterparties. *Source: Excerpt from IMF country report chapter on Austria (pages and figures as provided in the supplied content).* ### Annex II. Identification of Key Risk Factors: A Market-Based ### Annex II. Identification of Key Risk Factors: A Market-Based Approach ### Methodology - Time series: October 2007 to October 2012. - Market-based analysis using monthly changes in credit default swap (CDS) market and Moody’s KMV expected default frequencies (EDF) to measure changes in market perception of solvency risk. - For each bank, regress monthly changes in CDS spreads and EDF estimates on a set of explanatory variables covering: (i) Austria macro-financial variables; (ii) contagion from the CESEE region unrelated to domestic/global developments; and (iii) global risk factors, including changes in estimated risk premia. - Contagion from CESEE: compute monthly changes in sub-regional CDS weighted by consolidated BIS exposures of Austrian banks as of September 2008; regress each sub-region’s CDS changes on other explanatory variables and use orthogonal residuals as proxy for contagion from CESEE exposure. - Risk premia estimation: - Equity premium: monthly fluctuations in the price-earnings ratio (P/E) of the S&P 100 index. - Volatility risk premium: difference between implied (VIX) and realized volatility using the Garman and Klass (1980) efficient estimator; realized variance estimator denoted 2ˆσ computed as a 20-day period moving average on the S&P 100 index; volatility premium: VIX − 2ˆσ (denoted VIXpremVol). - Term premium: expected excess return on US 5 year Treasury bonds proxied by the linear combination of one through five year forward rates from Cochrane and Piazzesi (2005); changes proxied by change in expected average excess return of holding a two- through five-year government bond. - Data note: time span is driven by the availability of EDF estimates for a major bank. ### Variable set used in regressions - Dependent Variable: - monthly change in senior five year CDS quoted by CMA London (Bank 1) and 1-year EDF from Moody’s KMV (Bank 2). - Austrian Variables: - aut-return: return of the Austrian stock market from MSCI - gdp: monthly change in annual real GDP forecast for the current year - ip: monthly change in industrial production forecast for the current year - cab: monthly change in the current account balance forecast to GDP for the current year - w: monthly change in real wages forecast for the current year - u: the monthly change in the unemployment rate forecast for the current year - cds_aut: change in senior five year sovereign CDS for Austria - usd: change in the USD to euro - chf: change in the CHF to euro - recap: binary variable taking the value of 1 the month it received public recapitalization - CESEE Variables: - NMS-04: proxy of contagion from Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia - NMS-07: proxy of contagion from Bulgaria and Romania - SEE: proxy of contagion from Croatia - CIS: proxy of contagion from Kazakhstan, Russia and Ukraine - forint: change in the Hungarian forint to euro - leu: change in the Romanian leu to euro - Global Variables: - us_return: U.S. stock market excess return computed as the monthly value-weighted return on the NYSE, AMEX, and NASDA stock indices minus the one-month US T-bill return - ty: change in the five-year constant maturity US Treasury rate - liborois: change in the 3-month LIBOR and the OIS rate on the USD - itraxx: change in the itraxx europe senior financial index - itraxx_cross: change in the itraxx europe crossover index covering 40 CDS on the most liquid sub-investment grade european corporate entities - itraxx_eur: change in the itraxx europe index covering 125 CDS on investment grade european corporates - iips: change in the weighted sovereign CDS of Ireland, Italy, Portugal and Spain - eq-prem: change in the price-earnings ratio for the S&P 100 index - vol_prem: difference between the VIX index and a measure of realized volatility using the Garman-Klass (1990) efficient estimator for a 20-moving window on the S&P 100 index - term_prem: expected excess return on US 5 year Treasury using the methodology proposed by Cochrane and Piazzesi (2005) ### Key empirical findings - Main Austrian macroeconomic determinant: - An upward revision of unemployment forecast emerges as the main Austrian macroeconomic determinant of changes in solvency risk. - Contagion from CESEE: - Contagion from CESEE is the most notable risk factor. The explanatory power of the regression increases significantly from 0.1 to 0.5–0.6. - In terms of sub-regions, the effect is mainly driven by NMS-04 and NMS-07, with NMS-04 being associated with sharp depreciations of the Hungarian forint. - A negative outlook in the SEE and CIS region prove not to be statistically significant. - Global financial stress and credit: - Heightened stress in European banks measured by the Itraxx Europe senior financial index, and higher credit risk in sub-investment grade European corporates tend to widen solvency risk. - A pick-up of sovereign distress in Austria or in the GIIPS has no significant impact on Austrian banks’ solvency risk, consistent with limited exposure to domestic and European peripheral countries. - Term premium effects: - Results suggest that an increase in the term risk premium may contribute to rising credit spreads. - Mechanisms: by affecting movements in long-term rates (despite central banks’ monetary policy reaction) it may lift up banks’ funding costs (direct channel) and raise credit risk of households and nonfinancial corporations (indirect channel). *Source: Annex II. Identification of Key Risk Factors: A Market-Based Approach (IMF staff analysis).* ### Annex II. Table 2. Econometric Results of Risk Factors for Solvency Stress Test ### Annex II. Table 2. Econometric Results of Risk Factors for Solvency Stress Test ### Overview - Regression results reported for three regional groupings with two bank specifications each: Austrian (Bank 1, Bank 2), CESEE (Bank 1, Bank 2), Global (Bank 1, Bank 2). - Standard errors in parentheses. - Significance notation: *** p<0.01, ** p<0.05, * p<0.1. - Observations: 60 for every specification. ### Key econometric estimates (coefficients with standard errors) - market_return - Global Bank 1: 0.418 (0.618) - Global Bank 2: -0.008 (0.012) - chf - Austrian Bank 1: -1.374 (2.232) - Austrian Bank 2: -0.023 (0.041) - CESEE Bank 1: -1.730 (1.574) - CESEE Bank 2: -0.030 (0.025) - Global Bank 1: -0.093 (1.390) - Global Bank 2: -0.016 (0.023) - gdp - Austrian Bank 1: -3.633 (13.810) - Austrian Bank 2: -0.126 (0.255) - ip - Austrian Bank 1: 2.190 (5.003) - Austrian Bank 2: 0.051 (0.092) - w - Austrian Bank 1: -4.108 (4.042) - Austrian Bank 2: -0.088 (0.075) - u - Austrian Bank 1: -2.473 (9.595) - Austrian Bank 2: 0.266 (0.177) - CESEE Bank 1: 0.682 (6.819) - CESEE Bank 2: 0.277** (0.110) - Global Bank 1: -0.748 (5.760) - Global Bank 2: 0.131 (0.099) - cab - Austrian Bank 1: 18.266 (18.889) - Austrian Bank 2: -0.101 (0.349) - CESEE Bank 1: 10.743 (11.675) - CESEE Bank 2: 0.071 (0.191) - cds_aut - Austrian Bank 1: 0.317 (0.225) - Austrian Bank 2: -0.001 (0.004) - NMS-04 - CESEE Bank 1: 1.055*** (0.323) - CESEE Bank 2: -0.001 (0.005) - Global Bank 1: 0.268 (0.365) - Global Bank 2: -0.002 (0.006) - NMS-07 - CESEE Bank 1: -0.153 (0.192) - CESEE Bank 2: 0.008** (0.003) - Global Bank 1: -0.090 (0.175) - Global Bank 2: 0.009*** (0.003) - SEE - CESEE Bank 1: 0.191 (0.251) - CESEE Bank 2: -0.004 (0.004) - Global Bank 1: 0.231 (0.232) - Global Bank 2: -0.000 (0.004) - CIS - CESEE Bank 1: -0.033 (0.038) - CESEE Bank 2: 0.002*** (0.001) - Global Bank 1: 0.016 (0.045) - Global Bank 2: 0.000 (0.001) - forint (Global group) - Global Bank 1: 4.083*** (1.280) - Global Bank 2: -0.012 (0.020) - leu (Global group) - Global Bank 1: 2.359 (1.943) - Global Bank 2: 0.031 (0.033) - libor-ois (Global group) - Global Bank 1: -0.038 (0.157) - Global Bank 2: 0.010*** (0.003) - cds_iips (Global group) - Global Bank 1: 0.006 (0.005) - Global Bank 2: 0.000 (0.000) - recap (Global group) - Global Bank 1: -25.404 (30.396) - Global Bank 2: -0.439 (0.661) - itraxx (Global group) - Global Bank 1: 0.930*** (0.214) - Global Bank 2: 0.001 (0.004) - itraxx_cross (Global group) - Global Bank 1: 0.099** (0.044) - Global Bank 2: 0.004*** (0.001) - itraxx_eur (Global group) - Global Bank 1: -0.076 (0.256) - Global Bank 2: -0.004 (0.004) - eq_prem (Global group) - Global Bank 1: -0.066 (1.663) - Global Bank 2: 0.001 (0.030) - vol_prem (Global group) - Global Bank 1: -0.113 (0.639) - Global Bank 2: -0.005 (0.011) - term_prem (Global group) - Global Bank 1: 7.696 (11.273) - Global Bank 2: 0.446** (0.177) - Constant - Austrian Bank 1: 3.940 (6.009) - Austrian Bank 2: 0.037 (0.111) - CESEE Bank 1: 2.287 (4.326) - CESEE Bank 2: 0.007 (0.070) - Global Bank 1: 0.017 (3.751) - Global Bank 2: 0.031 (0.061) ### Model fit and sample - Observations: 60 for each specification. - R-squared - Austrian Bank 1: 0.108 - Austrian Bank 2: 0.087 - CESEE Bank 1: 0.503 - CESEE Bank 2: 0.610 - Global Bank 1: 0.766 - Global Bank 2: 0.813 ### Notes and significance summary - Standard errors reported in parentheses next to coefficients. - Significance markers present where applicable: *** p<0.01, ** p<0.05, * p<0.1. *AUSTRIA — INTERNATIONAL MONETARY FUND* ### Annex III. Stress Test Matrix (STeM) for the Banking Sector ### Annex III. Stress Test Matrix (STeM) for the Banking Sector ### Banking Sector: Solvency Risk - Institutional perimeter - Institutions included: Five largest banks (Bottom-Up); All banking institutions: approx. 585 consolidated (Top-Down). - Market share: Two thirds of banking sector assets (Bottom-Up); 100 percent of banking sector assets (Top-Down). - Data and baseline date: Institutions’ own data as of Q4 2012; Consolidated banking group (Bottom-Up). Supervisory data as of Q4 2012; Consolidated banking group (Top-Down). - Channels of risk propagation / Methodology - Banks’ internal models (Bottom-Up). - OeNB Balance-sheet model (Top-Down). - Satellite models: Internal models for market risk; separate satellite model for the Austrian portfolio (using insolvency data across six industry sectors) and the CESEE/CIS portfolio (using LLPR data broken down by currency), linking PDs/LGDs with macro scenarios. - Stressed PDs and LGDs; OeNB consensus rating across banks applied to single loans. - Solvency and funding interactions included. - Stress test horizon: Instantaneous (Bottom-Up); 2013–2015 (Top-Down). - Tail shocks / Scenario analysis - Global Slowdown/Euro Area Debt Crisis: Deviation of 2-year accumulated growth rate of 2.0 SD for the Austrian economy and 1.5 SD for the CESEE/CIS region. - Country specific projections developed for twenty-two CESEE/CIS countries. - Recession in the CESEE/CIS region: Country-specific add-on shocks to seven countries raising the overall size of the shock to 1.7 SD for CESEE and 1.8 for CIS. - Global Funding Scenario calibrated to Q3 2008 (domestic currency, Eurocurrency, deposit runs, FX swap markets, market issuances). - Sensitivity analysis - Market risk applied to trading book positions as of Dec 2012; valuation effects reported for a wide spectrum of stressed risk parameters. - Sovereign risk in both banking and trading book, including CESEE subsidiaries (government, regional, and local authorities) applied to 64 countries across all remaining maturities (3m, 6m, 1y-10y, 15y, 20y, 30y). - Credit risk from foreign currency lending: assumption: 1.5 SD of CHF for Austrian exposures and cross-border lending. - Funding risk for deposits and capital market issuance calibrated to 2008Q3-2009Q1. - Securitization portfolio following 2011 EBA methodology. - Market risk from underperformance of repayment vehicles and FX appreciation. - Risks and buffers - Risks/factors assessed: Market risk parameters: interest rates (23), major FX (8), equity indices (15), commodities (4), credit spreads (8), and counterparty risk (2). - Other assessed items: Credit losses, operating profits, funding costs, performance of repayment vehicles, sovereign risk, counterparty risk, exchange rate, taxes. - Full implementation of Basel III phase-in arrangements on aggregate CET1. - Behavioral adjustments: Constant balance sheet; No asset disposals allowed; No credit growth assumed. - Regulatory and market-based standards and parameters - Calibration: Major interest rates and FX rates jointly determined with the macro adverse scenario for consistency. - Use of EBA’s EU-wide stress test 2011, historical volatility, and expert judgment for remaining risk factors. - Stressed PDs and LGDs applied to compute both credit losses and stressed RWA calculations. - Capital definition according to EBA CT1 and Basel 2.5 RWAs. - Estimates of Basel III capital ratios (CET1, Tier 1, CAR) and RWAs for top 3, top 5, and whole banking system. - Hurdle rate: 5 percent, 6 percent, and 8 percent (CT1, T1, CAR) for whole banking system. Front-loaded CET1 ratio (7 percent) for large international banks. - Reporting / Output presentation - Dispersion of valuation losses: Absolute and in terms of capital. - Distribution of capital ratios. - Percentage of assets in capital buckets. - Weighted average capital ratios. - Percentage of assets that fail. - Recapitalization needs. ### Banking Sector: Liquidity Risk - Institutional perimeter - Institutions included: 29 banking institutions. - Market share: 80 percent of banking sector assets. - Data and baseline date: Supervisory data as of Q4 2012; Consolidated banking group. - Granular data based on contractual and behavioral expected cash-flows over five maturity buckets (5 days, 1m, 3m, 6m, and 12m). - Channels of risk propagation / Methodology - Cash-flow-based using 6 major currency buckets. - Scenarios based on the underlying macro-economic scenarios of the solvency stress test: (i) PD shifts feed into counterbalancing capacity and cash inflows; (ii) feedback effects included due to rising funding costs projected under the adverse macro scenario. - Risks and buffers - Risks: Impact of solvency on liquidity via three channels: NPL impact on cash inflows via a credit risk migration matrix; capital ratios on cash outflows (funding cost and rollover rates); asset quality on counterbalancing capacity. - Funding liquidity shock and market liquidity shock. - Buffers: Counterbalancing capacity taking into account haircuts to liquid assets. - Tail shocks / Size of the shock - Bank run and dry up of wholesale funding markets over 45 scenarios including full/limited/restricted/closed access to money markets covering funding in domestic currency, Eurocurrency funding, FX swap markets. - Scenarios grouped: baseline, market mild, market medium, market severe, and combined scenario (including market and idiosyncratic shocks). - Detailed assumptions and results reported for the mild market scenario consistent with recent EU FSAPs liquidity stress test scenarios. - Instantaneous outflow of funding and gradual outflow over 30-day, 90-day, and 1-year horizon. - Regulatory and market-based standards - Hurdle metrics: liquidity gap by major currency. - Definition of liquidity: local regulatory requirements. - Mapping with recent EU FSAP scenarios (stricter than revised Basel III LCR). - Reporting / Output presentation - Percentage of assets that fail under each horizon. ### Banking Sector: Contagion Risk - Institutional perimeter - Institutions included: Banks operating in the Austrian interbank market (Bottom-Up); Top listed Austrian banks and major global European banks (40) and in the CESEE (14) (Top-Down). - Market share: All banking institutions: 585 consolidated (Bottom-Up); Ranging between 40 percent and 60 percent of banking system assets (Top-Down). - Data and baseline date: Supervisory data as of Q4 2012 (Bottom-Up) — unconsolidated exposures but consolidated capital ratios; Balance sheet, market data as of Q4 2012 on consolidated basis (Top-Down). - Channels of risk propagation / Methodology - Impact study of liquidity risk on solvency risk. - Network analysis in the Austrian interbank market using a default cascade model. - CoVaR framework. - Asymmetric response in episodes of deleveraging. - State variables: a volatility index, a liquidity spread, changes in the short-end and the slope of the yield curve, changes in high-yield credit spreads, and equity market returns. - European and US financial variables considered. - Tail co-dependence assessed in (i) banks’ market valued assets’ growth rates, and (ii) banks’ equity returns. - Individual risk computed using (i) a quantile approach, and (ii) a GARCH (1,1) framework. - Tail shocks / Size of the shock - Contagion from negative funding gaps (linked to macro stress test) leading to: (i) fire sales of assets; (ii) rising funding costs; and (iii) partial closure of capital markets under a Global Funding Scenario replicating post-Lehman funding strains. - 5% quantile of the conditional loss distribution used as baseline tail event. - Robustness checks applied to 1% and 2.5% of the conditional loss distribution. - Reporting / Output presentation - Capital shortfall, system wide. - Impact on regulatory capital ratios. - Contribution to systemic risk. - Vulnerability to systemic risk. - Effect of financial state variables on tail inter-dependence. - Distribution of results for (i) a European banking system, and (ii) a CESEE banking system peer group. ### Annex IV. Sensitivity Analysis of Repayment Vehicle (RPV) Foreign Currency Loans — Key Findings and Statistics - Exposure and composition - In Q3 2012, 11.2 percent of net loans to the private sector were based on a repayment vehicle. - As of September 2012, RPV loans to domestic nonbanks amounted to €30.0 billion. - €27.3 billion (share of 20.7 percent) were granted to households. - €2.7 billion (share of 2.0 percent) to non-financial corporations. - RPV loans are bullet loans (repayment at maturity); monthly payments into an RPV during life of loan; maturity repayment from accumulated payments and returns. - Funding gaps and drivers - Latest survey in mid 2011: total funding gap of RPVs amounted to €5.3 billion (18.2 percent). - Mid-2011 funding gaps of RPV loans denominated in FC to domestic households and corporates amounted to €4.7 billion (with funding gaps of 20.1 percent and 18.6 percent for CHF and JPY, respectively). RPV loans denominated in Euro accounted for €0.6 billion (15.6 percent funding gap). - Mutual funds-based life insurance: - Account for 54.0 percent of RPVs. - Show the largest funding gap (21.0 percent against an average of 16.0 percent for remaining vehicles). - Contribution to the estimated aggregate shortfall as of September 2012 (€5.5 billion) is estimated at €3.4 billion. - Stress test sensitivity and results - FSAP team tail-risk sensitivity: bullet loans ‘at risk’ manageable even under extreme events. - Adding an additional annual yield shock (100 bps shock (adverse) and 200 bps shock (severe)) combined with a 1.5 SD CHF appreciation of the loan face value at maturity: - Maximum funding gap would increase from an accumulated €4.2 billion to €8.6 billion. - Given an average remaining maturity of 13 years, this implies an additional loss of under 1bn during the stress test horizon. - Caveats: extensive assumptions on breakdown of projected RPV value into assets already paid, future payments, and assumed performance of the RPV. Do not add these sensitivity results to sensitivity results in FCLs as that would duplicate FX shock effects. - Annex IV. Table 1 — selected aggregate figures (in million euros / percent as presented) - Total market sensitive: - Baseline: 17,224 - Adverse: 16,150 - Severe: 15,152 - Baseline (with 1.5 SD CHF): 21,483 - Adverse (with 1.5 SD CHF): 23,730 - Severe (with 1.5 SD CHF): 4,259 / 5,333 / 6,331 (projected funding gaps) and 6,506 / 7,580 / 8,578 (projected funding gap with yield and FX shock) depending on column pairing shown in table. - Percent funding gaps shown: 27.4%, 31.9%, 36.1%. - Total RPVs: 23,695 28,983 - Notes from table: - 1/ 50th percentile of daily annual returns over Jan 2005 through Dec 2012 used for proxy instruments. - 2/ Analysis assumes average residual maturity of 13 years as of September 2012. - 3/ An accumulated CHF appreciation of 1.5 SD over 2013-2014 with a stabilizing rate thereafter increases outstanding debt for CHF denominated loans. ### Annex V. Sovereign Risk Calibration — Methodology and Implementation - Scope and coverage - Sovereign scope follows CRD IV standardized approach: all central governments (but no central banks), all regional governments, and all local authorities. Exposures classified under IRB are segmented similarly. - All direct and indirect sovereign exposures stressed including those held by CESEE subsidiaries. - Net direct exposure: gross exposures (long) net of cash short position of sovereign debt (without derivative hedges such as CDS). - Indirect sovereign exposures include on and off balance sheet exposures. - Treatment of derivatives and indirect exposures - Direct derivatives positions: subject to fair value adjustments based on the relevant shock and relevant CVA adjustments. - Indirect exposures (counterparties other than the sovereign itself, i.e., CDS): treated similarly with fair value adjustments and CVA adjustment. - Calibration approach - Under stress, the term structure of sovereign risk shifts upward for all countries to which Austrian banks are exposed, including sovereign bonds held by CESEE subsidiaries to comply with local liquidity requirements. - Approach allows changes in risk term premia associated with excess yield investors require for holding long-term bonds under volatile conditions. - Calibration of sovereign shock on the level of spreads (parallel shift), slope, and curvature of the yield curve based on historical yields at each maturity, using modified duration approach. When no available maturity exists to derive a valuation haircut, haircuts are interpolated. - Numerical calibration specifics - Shock calibrated for fifty eight countries. - For fifty countries, shock derived from the 90th percentile of the historical distribution of annual changes of daily yields of Bloomberg generic 5-year government bond yields over the period 2005–12. - For Belarus, Bulgaria, Luxembourg, Malta, and Romania, haircut computed using extreme returns for the most liquid outstanding international bond as of Dec 2012 due to limited generic yield curve time series. - For Cyprus, haircut calibrated from the sovereign yield curve as of Dec 2012. - For Estonia, only international loans were outstanding as of Dec 2012. - Application to banking book - Haircuts to the banking book applied to adjusted (marked-to-market) balance sheet values — banks recognized gains/losses before the haircut from search-for-yield or flight-to-quality dynamics. - Exposures reported before deduction of provisions, application of credit conversion factors, or credit risk mitigation techniques. - For exposures valued at amortized cost, valuation loss for each country and sovereign bucket calculated as: - Valuation loss = amortized cost - market value + market value * haircut. - Resulting losses distributed across the stress testing horizon. *Source: Annex III–V, _cr1416 - Annex III. Stress Test Matrix (STeM) for the Banking Sector (IMF staff extract)._ ### Annex V. Table 2. Sovereign Haircuts ### Annex V. Table 2. Sovereign Haircuts ### Sovereign haircuts by country and maturity (key observations) - The table reports sovereign haircuts (in percent) by selected countries of exposure across maturities: 3M, 6M, 1Y, 2Y, 3Y, 4Y, 5Y, 6Y, 7Y, 8Y, 9Y, 10Y, 15Y, 20Y, 30Y (values shown where available). - Extremely large haircuts are reported for Greece (GR) at selected maturities: - 1Y: 125.29 - 2Y: 168.69 - (further entries include) 293.09 - Large sovereign haircuts for Portugal (PT) at long maturities: - 2Y: 16.15 - 3Y: 26.19 - 4Y: 28.42 - 5Y: 37.93 - 6Y: 42.97 - 7Y: 47.77 - 8Y: 52.69 - 9Y: 54.12 - 10Y: 47.42 - 30Y: 87.76 - Slovenia (SI) shows high haircuts at longer maturities: - 2Y: 0.02 - 4Y: 6.22 - 5Y: 4.51 - 6Y: 11.10 - 7Y: 10.88 - 8Y: 16.25 - 9Y: 14.72 - 10Y: 21.46 - 15Y: 24.05 - 20Y: 19.33 - 30Y: 41.59 - Hungary (HU) displays elevated haircuts across maturities: - 3M: 2.00 - 6M: 3.00 - 1Y: 4.00 - 2Y: 5.00 - 3Y: 7.42 - 4Y: 9.78 - 5Y: 11.50 - 6Y: 15.00 - 7Y: 20.79 - 8Y: 22.50 - 9Y: 21.00 - 10Y: 20.70 - 15Y: 28.03 - Ireland (IE) shows large haircuts at longer maturities (values listed where shown): - (entries include) 2.27, 7.90, 12.74, 23.71, 20.36, 16.73, 26.54, 28.00, 30.00, 33.00 - Spain (ES) increasing with maturity: - 3M: 2.96 - 6M: 4.28 - 1Y: 5.73 - 2Y: 7.13 - 3Y: 8.01 - 4Y: 9.04 - 5Y: 10.05 - 6Y: 11.49 - 7Y: 12.59 - 8Y: 18.44 - 9Y: 25.71 - 10Y: 31.96 - Italy (IT) profile across maturities: - 3M: 0.29 - 6M: 0.58 - 1Y: 1.18 - 2Y: 2.95 - 3Y: 4.35 - 4Y: 5.25 - 5Y: 6.42 - 6Y: 7.32 - 7Y: 8.20 - 8Y: 8.59 - 9Y: 8.91 - 10Y: 10.08 - 15Y: 13.27 - 20Y: 18.75 - 30Y: 23.09 - Germany (DE) low haircuts at short maturities, modestly increasing: - 3M: 0.28 - 6M: 0.56 - 1Y: 1.15 - 2Y: 2.14 - 3Y: 3.03 - 4Y: 3.75 - 5Y: 4.11 - 6Y: 4.55 - 7Y: 4.91 - 8Y: 5.12 - 9Y: 5.41 - 10Y: 5.61 - 20Y: 10.34 - 30Y: 13.68 - United Kingdom (GB) term structure: - 3M: 0.03 - 6M: 0.05 - 1Y: 0.82 - 2Y: 1.55 - 3Y: 2.17 - 4Y: 2.71 - 5Y: 3.13 - 6Y: 3.44 - 7Y: 4.51 - 8Y: 5.05 - 9Y: 6.84 - 10Y: 5.68 - 15Y: 6.78 - 20Y: 8.04 - 30Y: 11.37 - United States (US) term structure (selected maturities): - 3M: 0.37 - 6M: 0.67 - 1Y: 0.06 - 2Y: 1.62 - 3Y: 2.12 - 4Y: 2.12 - 5Y: 3.56 - 6Y: 2.12 - 7Y: 2.00 - 8Y: 4.00 - 9Y: 6.67 - 10Y: 9.00 - 15Y: 12.00 - 20Y: 17.78 - Japan (JP) very low haircuts across maturities: - 3M: 0.00 - 6M: 0.00 - 1Y: 0.01 - 2Y: 0.02 - 3Y: 0.08 - 4Y: 0.16 - 5Y: 0.10 - 6Y: 0.08 - 7Y: 0.19 - 8Y: 0.26 - 9Y: 0.27 - 10Y: 0.25 - 15Y: 1.29 - 20Y: 2.28 - 30Y: 5.10 - Several other countries and country groups show varied profiles (representative entries): - Austria (AT): 3M 0.28; 6M 0.56; 1Y 0.82; 2Y 2.18; 3Y 3.17; 4Y 3.68; 5Y 4.48; 6Y 4.33; 7Y 5.16; 8Y 5.25; 9Y 6.64; 10Y 6.24; 15Y 8.19; 20Y 19.25 - Poland (PL): 3M 0.28; 6M 0.56; 1Y 1.63; 2Y 2.52; 3Y 3.11; 4Y 3.54; 5Y 4.02; 6Y 1.55; 7Y 2.41; 8Y 3.82; 9Y 3.18; 10Y 7.28; 15Y 10.04 - Romania (RO): 3M 0.27; 6M 0.24; 1Y 1.06; 2Y 2.13; 3Y 3.19; 4Y 4.25; 5Y 5.31; 6Y 6.00; 7Y 7.00; 8Y 8.00; 9Y 9.00; 10Y 10.63; 15Y 15.94 - Russia (RU): 3M 0.50; 6M 0.90; 1Y 1.61; 2Y 2.75; 3Y 5.61; 4Y 3.12; 5Y 4.30; 6Y 4.44; 7Y 5.00; 8Y 6.00; 9Y 7.00; 10Y 7.19; 15Y 8.00 - Turkey (TR): 3M 2.74; 6M 5.24; 1Y 6.48; 2Y 7.65; 3Y 7.83; 4Y 8.00; 5Y 8.50; 6Y 9.00; 7Y 9.50; 8Y 10.05 - Belgium (BE): 3M 2.80; 6M 4.17; 1Y 5.09; 2Y 5.20; 3Y 5.81; 4Y 6.28; 5Y 6.49; 6Y 6.91; 7Y 7.13; 9Y 13.26; 10Y 11.36; 15Y 18.83 - Cyprus (CY): 3M 11.68; 6M 20.26; 1Y 36.03; 2Y 50.90 - Finland (FI): 3M 0.00; 6M 0.00; 1Y 0.05; 2Y 0.68; 3Y 1.23; 4Y 1.28; 5Y 2.12; 6Y 3.00; 7Y 3.86; (other entries include) 0.00, 1.00, 1.24, 3.18 - Luxembourg (LU): 3M 0.15; 6M 0.31; 1Y 0.59; 2Y 2.17; 3Y 3.10; 4Y 3.72; 5Y 4.20; 6Y 4.73; 7Y 5.05; 8Y 5.30; 9Y 5.50; 10Y 5.91; 15Y 6.12; 20Y 11.20; 30Y 14.09 - Netherlands (NL): same values as Luxembourg for listed maturities: 3M 0.15; 6M 0.31; 1Y 0.59; 2Y 2.17; 3Y 3.10; 4Y 3.72; 5Y 4.20; 6Y 4.73; 7Y 5.05; 8Y 5.30; 9Y 5.50; 10Y 5.91; 15Y 6.12; 20Y 11.20; 30Y 14.09 - Switzerland (CH): entries include 6M 0.23; 1Y 0.46; 2Y 0.10; 3Y 0.51; 4Y 0.45; 5Y 0.59; 6Y 1.19; 7Y 1.52; 8Y 1.12; 9Y 3.16; 10Y 3.38; 15Y 4.41 - Other RoW and Other SEE and BYOther CISHR/Other SEE groups show repeated patterns: 3M 2.00; 6M 3.00; 1Y 4.00; 2Y 5.00; 3Y 6.00; 4Y 6.68; 5Y 6.94; 6Y 5.06; 7Y 8.00; 8Y 11.17; 9Y 13.00; 10Y 16.00; 15Y 18.00 - Several entries in the table are elided with "..." where values are not shown in the excerpt. ### Methodology and notes (explicit) - Source: Bloomberg and IMF staff estimates. - The shock is derived from the 90th percentile of the historical distribution of annual changes of daily yields of Bloomberg generic 5-year government bond yields over the period 2005–12. - For Belarus, Bulgaria, Luxembourg, Malta, and Romania, the haircut is computed using extreme returns for the most liquid outstanding international bond as of Dec 2012. - For Cyprus, the haircut is calibrated from the sovereign yield curve as of Dec 2012. *Source: Bloomberg and IMF staff estimates (Annex V. Table 2, _cr1416)._ ### Annex VI. Table 6. Inward Spillovers from CESEE Peer Banks ### Annex VI. Table 6. Inward Spillovers from CESEE Peer Banks ### Overview - Metric: Weekly average, in percent of market valued asset returns. - Definition: Contribution to systemic risk is defined as the average pair wise increase in distress of each vulnerable bank when the contributor bank is in distress relative to each vulnerable bank's unconditional VaR at the 5th percentile. - Source: IMF staff calculations. ### Key statistics — AUSTRIA (pairwise entries shown as "Contributor-Value Vulnerability" and "Contributor-Value Contribution") - OTP-13.0 - KBC-11.1 - KBC-11.1OTP-9.1 - ACA-8.3BNP-7.3 - GLE-7.4SAN-6.7 - BNP-7.0UCG-6.4 - SEBA-6.4ACA-6.4 - UCG-6.3GLE-6.4 - SAN-5.7INGA-5.9 - INGA-4.4SEBA-5.8 - ISP-3.8SEWDA-5.8 - SWEDA-3.2ISP-3.9 - CBK-1.1CBK-1.5 - Average-6.5Average-6.4 (Note: entries are presented exactly as in the source text.) ### Annex VI. Figure 2 — CoVaR: Individual vs. Systemic Risk - Scope: European Banking System and CESEE Banking System. - Measure of solvency risk: VaR. - Measure of contribution to systemic risk: ∆ CoVaR. - Sample period for both measures: averages of weekly market valued returns over 2005-2012. - Key takeaway from figure note: The scatter plot shows the weak link between banks' solvency risk (VaR) and their contribution to systemic risk (∆ CoVaR). ### Notes and definitions - "Average vulnerability of Austrian banks to CESEE peers' bilateral distress." (label 1/ in source) - "Average bilateral contribution to distress across CESEE peers." (label 2/ in source) ### References (as listed in source) - Adrian, T. and M. Brunnermeier, 2011, “CoVaR”, Working paper, Princeton University. - Albacete, N., and Fessler, P., 2010, “Stress Testing Austrian Households”, Financial Stability Report 19, June, OeNB, pp. 83-95. - Boss, M., G. Krenn, C. Puhr, and M. Summer, 2006, “Systemic Risk Monitor: A Model for Systemic Risk Analysis and Stress Testing of Banking Systems”, Financial Stability Report 11, December, OeNB, pp. 72-87. - Cochrane, J. H., and M. Piazzesi, 2005, “Bond Risk Premia”, American Economic Review, 95(1)”, pp. 138-60/ - European Banking Authority (EBA), 2011, “2011 EU-Wide Stress Test: Methodological Note—Version 1.1,” March 18. - Financial Stability Oversight Council, 2013, “A Framework to Mitigate Systemic Risk”, CRS Report for Congress. - Furfine, C., 2003, “Interbank Exposures: Quantifying the Risk of Contagion”, Journal of Money, Credit and Banking 35 (1), pp. 111-128. - Garman, M. B., and M. J. Klass, 1980, “On the Estimation of Security Price Volatilities from Historical Data”, Journal of Business, 53 (1), pp. 67-78. - International Monetary Fund, 2009, Global Financial Stability Report, Chapter 2 (October) (Washington: International Monetary Fund). - International Monetary Fund, 2012, World Economic Outlook, Chapter 1 (October) (Washington: International Monetary Fund). - International Monetary Fund, 2013, Fiscal Monitor, Chapter 1 (April) (Washington: International Monetary Fund). - International Monetary Fund, 2013, World Economic Outlook, Chapter 1 (April) (Washington: International Monetary Fund). - International Monetary Fund, 2013, Central, Eastern and Southeastern Europe –Regional Economic Issues, (April) (Washington: International Monetary Fund). - International Monetary Fund, 2013, Austria: Financial Stability Assessment, IMF Country Report No. 13/X (Washington). - Kerbl, S. and M. Sigmund, 2011, “What Drives Aggregate Credit Risk?”, Financial Stability Report 22, December, OeNB, pp. 72-87. - Liikanen, E. 2012, “High-level Expert Group on reforming the structure of the EU banking sector”, October. - Longstaff, F., J. Pan, L. H. Pedersen, and K. J. Singleton, 2011, “How Sovereign is Sovereign Credit Risk?”, American Economic Journal: Macroeconomics 3 (April 2011), pp. 75-103. - López-Espinosa, G, A. Moreno, A. Rubia, and L. Valderrama, 2012, “Short-term wholesale funding and systemic risk: A global CoVaR approach”, Journal of Banking and Finance, vol. 36(12), pp. 3150-3162. - Oesterreichische Nationalbank, 2012, Financial Stability Report, December, OeNB. - Puhr, C. and S. W. Schmitz, 2013, “A view from the top – the interaction between solvency and liquidity stress”, OeNB mimeo. - Schneider, M. and M. Leibrecht, 2006, “AQM-06: The Macroeconomic Model of the OeNB”, OeNB Working Paper No. 132. - Schmitz, S. W., and F. Weidenholzer, 2009, “Recent Developments in the Austrian Banking System’s Liquidity Situation and the International Regulatory Debate,” Financial Stability Report 18, December, OeNB, pp. 60-66. - Tarullo, D. K, 2011, “Regulating Systemic Risk,” Speech, 2011 Credit Markets Symposium, North Carolina, Charlotte, March 31, 2011, Board of Governors of the Federal Reserve System. *Source: IMF staff calculations.* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2014/_cr1416.pdf_