## _cr13185 - Executive Summary ## Source details **Canonical URL:** [_cr13185 - Executive Summary](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13185.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13185.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13185.pdf.json) --- ### Executive summary: key findings - Stress testing analysis supports ACP requirement for banks to build up adequate capital and liquidity buffers. - Banking system would be able to meet regulatory ratios under most scenarios, but pockets of vulnerabilities need addressing, in particular regarding liquidity risks. - Solvency stress tests indicate banks could cope with deterioration in the economic environment while phasing in capital requirements under Capital Requirements Directive (CRD) IV. - BU (bottom up) tests were the core element and were cross-validated by TD (top down) tests; TD results were more macro-sensitive and characterized by lower Common Equity Tier 1 (CET1) ratios than banks’ results. - The adverse scenario has its largest impact in 2012 and 2013; changes in capital ratios in 2012 and 2013 are largely driven by credit risk. - Recovery phase during 2014–16 helps smooth the impact of the introduction of CRD IV; largest regulatory-impact reductions in CET1 ratio estimated at about 43 bps in 2015 and 82 bps in 2016 (above the impact of the adverse scenario). - BU stress tests indicate shocks to equity and real estate prices have a negligible impact on CET1 ratios. - Exposures to sovereigns in the European periphery were cut substantially in the second half of 2011: cross-border public sector exposures fell by 38 percent to Italy, 39 percent to Spain, 39 percent to Greece, 32 percent to Portugal, and 26 percent to Ireland. - Available-for-sale (AFS) and trading account holdings of non-AAA sovereign debt remain sizeable, particularly to Italy; a worsening of the euro crisis would cause losses of about 5 percent of the initial CET1 capital level. Prudential filters (a cumulative 20 percent each year starting in 2014) would phase in capital charges over the risk horizon. - An extreme shock affecting all sovereign holdings (including France) in all books would impact the initial aggregate CET1 capital level by an additional 5 percent. ### System structure and background - France’s financial system is large, sophisticated, and diversified and is dominated by five banking groups that are regionally and globally systemic; four have been identified as global systemically important banks (G-SIBs). - Total assets of the system amounted to about six times France’s GDP at end-2011. - Rapid balance sheet growth during the second half of the 2000s was driven by expansion into international corporate and investment banking (CIB) and derivatives products, funded in the wholesale market. - Deleveraging plans were announced by the five largest banks in mid-2011, focused on disposing of noncore, dollar-funded international assets. - Stress testing exercise covered eight banking groups accounting for 97 percent of banking assets: BNP Paribas, 27 percent; Groupe Crédit Agricole, 24 percent; Société Générale, 16 percent; Groupe BPCE, 16 percent; Groupe Crédit Mutuel, 7 percent; HSBC France, 3 percent; La Banque Postale, 3 percent; Caisse des Dépôts et Consignations, 2 percent. (Exercise included on-balance sheet part of Caisse des Dépôts et Consignations only.) ### Solvency stress tests: methodology and outcomes - Coverage and timing: - BU tests by banks and TD tests by authorities on portfolios as of end-December 2011; analysis at consolidated group level. - Risk horizon: five years (2012–2016); market and sovereign risks and sensitivity tests assessed instantaneous impact as of December 2011. - Components: - Single-factor stress tests and macroeconomic scenario analysis; assessed valuation of assets, liabilities, and profits; focused on credit, market, concentration, liquidity, and contagion risks. - Banks translated macro variables into income, expense, loan loss and capital requirement forecasts; loan loss forecasts disaggregated into Regulatory PD and downturn LGD. - ACP TD models: (1) Return on asset ratio regression to calculate stressed capital level; (2) Transition matrices model to calculate stressed RWA. - Key assumptions: - Balance sheets assumed to grow with nominal GDP. - Funding structure and hedging strategy assumed unchanged; maturing liabilities replaced by similar ones. - Banks used through-the-cycle (TTC) PDs for RWA; BU repricing of non-AAA sovereigns limited to trading and AFS accounts; TD did not include sovereign risk explicitly. - Future regulatory changes captured only if they came into force during 2012–2016; phase-in effects taken into account. - Solvency thresholds (In percent): - Forecast Year: 2012 2013 2014 2015 2016 - Minimum total capital: 8.0 8.0 8.0 8.0 8.0 - Minimum tier 1 capital: 4.0 4.5 5.5 6.0 6.0 - Minimum common equity tier 1: 2.0 3.5 4.0 4.5 4.5 - Conservation Buffer: 0.0 0.0 0.0 0.0 0.625 - Additional Loss Absorbency for G-SIBs: 0.0 0.0 0.0 0.0 0.5 - Results and drivers: - Aggregate conclusion: system resilient to wide range of adverse shocks but pockets of vulnerabilities exist. - TD results more macro-sensitive and showed lower CET1 ratios than BU results due to differences in assumptions (e.g., more conservative and cyclical TTC PDs for non-retail claims used by authorities). - Adverse scenario largest impact in 2012 and 2013; credit risk and higher provisions are main drivers in those years. - RWA changes limited by mild deleveraging (credit growth follows nominal GDP, projected to decline by 0.7 percent in 2012) and by use of through-the-cycle PDs. - Authorities estimate net regulatory-impact reductions in CET1 ratio of about 43 bps in 2015 and 82 bps in 2016 above the adverse scenario impact. ### Solvency stress tests: scenarios, shocks, and satellite models - Macro-scenario framework: - Two scenarios: baseline and adverse; baseline follows February 2012 WEO update. - Adverse scenario: cumulative deviation from baseline of 2.1 standard deviation of GDP growth for 2012–13. - Adverse scenario combined four adverse shocks: 1. Reduction in external demand from a global recession starting in 2012Q1. 2. Worsening of the European debt crisis: sovereign spreads up by 90 bps for France and 160bps for the euro area; fiscal consolidation to reduce fiscal deficit to 3 percent of GDP by 2013. 3. Worsening in banking funding costs leading to a credit contraction of -0.8 percent in 2012 (in addition to deleveraging plans). 4. Increase in sovereign risk causing repricing of bond holdings (through haircuts); in BU tests affected non-AAA sovereigns in trading and AFS accounts. - Selected macroeconomic path (Real GDP growth, Annual percentage change): - 2011 2012 2013 2014 2015 2016 - Baseline: 1.7 0.5 1.0 1.8 1.9 1.9 - Adverse: 1.7 -1.8 0.2 2.4 2.5 2.3 - Market and sensitivity tests included: - Instantaneous shocks: liquidation of non-AAA sovereign bonds in HTM; failure of largest corporate exposures; FX shock +/-20 percent; interest rate shock of 200 bps; 25 percent real estate price shock; reverse liquidity test of wholesale funding loss. - Sovereign risk measurement (BU): - Baseline shock derived from 50th percentile of historical annual changes of daily yields; adverse shock from 90th percentile. - Haircuts not applied to AAA countries; haircuts concentrated in first year of the adverse scenario. - Phase-out of AFS regulatory filter (percent): 2011: 0; 2012: 0; 2013: 0; 2014: 20; 2015: 40; 2016: 60. ### Liquidity stress tests: methodology and outcomes - Design: - BU liquidity stress tests used same eight banks at end-December 2011; run-off rates and haircuts calibrated by type using banks’ historical crisis data. - Two scenarios: without and with access to ECB facilities. - Maturity ladder analysis capturing outflows, standby inflows, and available liquidity buffers; survival horizon up to two years measured via residual stressed funding gap after asset sales. - Key findings: - Without recourse to ECB liquidity: - Two banks would face difficulties meeting liquidity needs from outflows with available buffers, standby inflows, and asset sales. - One bank cannot meet contractual obligations at one day to one week. - One bank cannot meet contractual obligations from three months to six months. - A two-notch bank downgrade could impose added stress through collateral and margin calls with significant effects on some banks. - With ECB support: - All banks would pass the liquidity test assuming access to ECB liquidity. - Reverse stress test on maximum potential loss of wholesale funding by currency: - Three of the banks recurring to central bank liquidity above a 5 percent loss of wholesale funding (in the event of closure of funding markets). - With ECB support, four banks would be able to address up to a maximum loss of about 15 percent of all wholesale funding. - Survival-period results (number of banks which still meet contractual obligations): - Without ECB support: 7, 6, 6, 6, 6, 5, 5, 5 (corresponding to buckets from Up to 1 day through Greater than 1 year up to 2 years). - With ECB support: 7, 7, 7, 7, 7, 7, 7, 7. ### Contagion risk: methodology and findings - Network stress-test approaches: - Direct network stress test using net exposures (Cont et al. (2010)); contagion begins when net exposure losses drive a bank below regulatory capital fixed at 7.0 percent of RWA. - Reverse network stress test using bilateral exposures (Gouriéroux, Héam, and Monfort (2012)) defining initial shock as loss on non-interbank assets triggering another bank’s default. - Network characteristics: - French banking network moderately concentrated; most interbank exposures relatively small. - In net exposures: two banks lend 60 percent of all interbank net flows and one bank receives over 40 percent of French interbank flows; these exposures are relatively small (under 3 percent of total assets). - Results (as of December 2011): - Failure of a single bank would result in a CET1 ratio decline to 8.5 percent from 9.9 percent. - Combined direct and reverse methods: over seven scenarios (failure of a single bank), only three scenarios led to failure of other banks (capital ratio below 7 percent) due to contagion. - Worst-case initial shock (first technique): system capital ratio moves from 9.9 percent to 9.1 percent due to contagion effects. - Second technique: no bank loss would trigger default of another bank; as a reverse stress test, contagion risk within French banking system was null. - Worst-case French banking system capital ratio reported: 8.48 percent, representing a loss of 140 basis points. - External contagion: - Larger contagion effects may emerge from exposures to non-French bank counterparts (other European banks, U.S. investment banks, banks from Japan and GCC countries). - Euro money market topology is highly concentrated with few, highly connected core players. ### Recommendations on stress testing - ACP to undertake frequent “bank-by-bank” solvency and liquidity stress tests of the largest French banks (including subsidiaries) and make results publicly available in the Financial Stability Report. - Priority: High. Timeframe: Immediate. - ACP to enhance current approaches to TD stress tests by introducing methodologies that can project risk losses by type or risk. - Priority: High. Timeframe: Immediate. - ACP to use more granular bank-by-bank data for calculation of risk parameters related to retail lending in TD stress testing. - Priority: High. Timeframe: Immediate. - Note on timing nomenclature: “Immediate” is within one year; “near-term” is 1–3 years; “medium-term” is 3–5 years. - Additional recommendations: - Liquidity stress tests should be undertaken frequently. - Both solvency and liquidity results should be made publicly available in the Financial Stability Report. - Strengthen ACP’s TD stress testing approaches and collect necessary data (e.g., for retail lending risk parameters). ### Additional technical and disclosure notes - Tests relied primarily on BU stress tests and were cross-validated by TD tests; owing to insufficient public disclosure and legal constraints, the FSAP team conducted a partial TD exercise using publicly available information, estimated RWA based on point-in-time PDs, and applied full Basel III with no filters on AFS positions. - Partial TD exercise broadly supported system resilience conclusion but illustrated sensitivity to assumptions, including PDs and use of capital measures as defined under CRD IV. - Exercise underscored importance of improving public disclosure of French financial institutions. ### Sovereign shocks, haircuts, and Appendix II sample values - AFS filter phase-in (percent): 2011: 0; 2012: 0; 2013: 0; 2014: 20; 2015: 40; 2016: 60. - Selected tentative haircuts and change in yields (50th and 90th percentiles) — selected entries as provided: - France: Tentative Haircut 2.10 90th Pctile Change in Yield 90.40 50th Pctile Change 0.76 90th Pctile Change 4.16 (see Appendix II and Appendix Table 4/5 excerpts for full country list and precise columns). - Italy: Tentative Haircut 33.40 90th Pctile Change in Yield 127.20 50th Pctile Change 1.67 90th Pctile Change 6.36. - Spain: Tentative Haircut 34.00 90th Pctile Change in Yield 149.20 50th Pctile Change 1.70 90th Pctile Change 7.46. - Greece: Tentative Haircut 72.80 90th Pctile Change in Yield 869.50 50th Pctile Change 3.64 90th Pctile Change 43.44. - Debt-weighted averages: - Euro Area: Tentative Haircut 14.87 90th Pctile Change in Yield 162.05 50th Pctile Change 0.74 90th Pctile Change 8.10. - All countries: Tentative Haircut -10.77 90th Pctile Change in Yield 90.43 50th Pctile Change -0.54 90th Pctile Change 4.52. - Impact estimates: - Worsening of euro crisis would cause cumulated losses of capital of about 5 percent of initial CET1 capital. - An extreme shock affecting all sovereign holdings (including France) in all books would impact the initial aggregate CET1 capital by an additional 5 percent. ### Housing: assessment and risks - Staff assessment: Medium; Staff assessment (alternate line): Medium-Low. - Drivers: housing prices supported by fundamentals, asset allocation factors (real estate as safe haven), and low interest rates. - Valuation: Staff estimates France’s housing prices were 10–25 percent overvalued at end-2010. - Downside risks: high LTV ratios, worsening economic outlook, possible unexpected increase in interest rates. - Risks to banks: appear limited given comparatively low household debt and overall sound lending standards; indirect effects possible via wealth and confidence channels. *Source: IMF staff summary of stress-testing exercise, based on analysis from Banque de France and ACP.* ### Executive Summary ...................................................................................................... ### _cr13185 - Executive Summary ### Executive summary: key findings - Stress testing analysis was used to capture the most salient risks for banks and supports the current focus of Autorité de Contrôle Prudentiel (ACP) to require banks to build up adequate capital and liquidity buffers. - The banking system would be able to meet regulatory ratios under most scenarios, but there are pockets of vulnerabilities that need to be addressed, in particular regarding liquidity risks. - Solvency stress tests indicate banks could cope with deterioration in the economic environment while phasing in capital requirements under Capital Requirements Directive (CRD) IV. - BU (bottom up) tests were the core element and were cross-validated by TD (top down) tests; TD results were more macro-sensitive and characterized by lower Common Equity Tier 1 (CET1) ratios than banks’ results. - The adverse scenario has its largest impact in 2012 and 2013. Changes in capital ratios in 2012 and 2013 are largely driven by credit risk. - The recovery phase during 2014–16 helps smooth out the impact of the introduction of CRD IV; the largest impact from the introduction of the new regulation takes place in 2015 and 2016 (a reduction in CET1 ratio of about 43 bps and 82 bps, respectively, above the impact of the adverse scenario). - BU stress tests indicate shocks to equity and real estate prices have a negligible impact on CET1 ratios. - Exposures to sovereigns in the European periphery were cut substantially in the second half of 2011 and have reduced French banks’ vulnerability to a sovereign shock. In the second half of 2011, cross-border public sector exposures of French banks fell by 38 percent to Italy, 39 percent to Spain, 39 percent to Greece, 32 percent to Portugal, and 26 percent to Ireland. - Available-for-sale (AFS) and trading account holdings of non-AAA sovereign debt remain sizeable, in particular to Italy; a worsening of the euro crisis would cause losses of about 5 percent of the initial CET1 capital level. Prudential filters (a cumulative 20 percent each year starting in 2014) would phase in capital charges over the risk horizon. - An extreme shock affecting all sovereign holdings (including France) in all books would impact the initial aggregate CET1 capital level by an additional 5 percent. ### System structure and background - France’s financial system is large, sophisticated, and diversified and is dominated by five banking groups that are regionally and globally systemic; four have been identified as global systemically important banks (G-SIBs). - Total assets of the system amounted to about six times France’s GDP at end-2011. - Rapid balance sheet growth during the second half of the 2000s was driven by expansion into international corporate and investment banking (CIB) and derivatives products, funded in the wholesale market. - Deleveraging plans were announced by the five largest banks in mid-2011, focused on disposing of noncore, dollar-funded international assets. - The stress testing exercise covered the major eight banking groups accounting for 97 percent of banking assets: BNP Paribas, 27 percent; Groupe Crédit Agricole, 24 percent; Société Générale, 16 percent; Groupe BPCE, 16 percent; Groupe Crédit Mutuel, 7 percent; HSBC France, 3 percent; La Banque Postale, 3 percent; and Caisse des Dépôts et Consignations, 2 percent. (The exercise included the on-balance sheet part of the Caisse des Dépôts et Consignations only.) ### Solvency stress tests: methodology and outcomes - Coverage: largest French banks; BU tests by banks and TD tests by authorities on portfolios as of end-December 2011; analysis at consolidated group level. - Tests comprised single-factor stress tests and macroeconomic scenario analysis; assessed valuation of bank assets, liabilities, and profits; focused on credit, market, concentration, liquidity, and contagion risks. - Credit risk under the adverse scenario raised probabilities of default for corporate and retail customers, forcing higher provisions; changes in risk-weighted assets (RWA) were limited by a mild deleveraging effect embedded in the scenario (credit growth follows nominal Gross Domestic Product (GDP), which is projected to decline by 0.7 percent in 2012) and by the use of through-the-cycle probabilities of default (PDs). - Sensitivity tests of concentration point to predominance of credit risk from name concentrations in France, Italy, and the United States. - The largest impact from introduction of CRD IV occurs in 2015 and 2016: reduction in CET1 ratio of about 43 bps in 2015 and 82 bps in 2016 (above the impact of the adverse scenario). ### Liquidity stress tests: methodology and outcomes - Liquidity stress tests assessed resilience to a strong shock characterized by run-off rates and haircuts on assets calibrated by type on French historical data, and assumed no market access. - Assuming no recourse to European Central Bank (ECB) liquidity, significant reliance on short-term funding would result in difficulties for two banks to meet liquidity needs from outflows with available buffers, standby liquidity from inflows, and asset sales. - A two-notch bank downgrade under these circumstances could impose added stress through collateral and margin calls with a significant effect on some banks. - All banks would pass the liquidity test assuming access to ECB liquidity. - A reverse stress test on the maximum potential loss of wholesale funding, by currency, that each bank could suffer while still meeting contractual obligations shows dependence on ECB funding in the event of a closure of funding markets, with three of the banks recurring to central bank liquidity above a 5 percent loss of wholesale funding. - With ECB support, four banks would be able to address up to a maximum loss of about 15 percent of all wholesale funding. ### Contagion risk - The French banking network is moderately concentrated, with most interbank exposures within it relatively small. - In terms of net exposures, two banks lend 60 percent of all interbank net flows and one bank receives over 40 percent of French interbank flows; these exposures are relatively small (under 3 percent of total assets). - Contagion risk among French banks appears limited; failure of a single bank would result in a CET1 ratio decline to 8.5 percent from 9.9 percent. - Larger contagion effects may emerge from exposures to non-French bank counterparts in the interbank market, including other European banks, U.S. investment banks, and banks from Japan and Gulf Cooperation Council (GCC) countries. ### Recommendations on stress testing (Table 1 highlights) - ACP to undertake frequent “bank-by-bank” solvency and liquidity stress tests of the largest French banks (including subsidiaries) and made results publicly available in the Financial Stability Report. Priority: High. Timeframe: Immediate. - ACP to enhance current approaches to TD stress tests by introducing methodologies that can project risk losses by type or risk. Priority: High. Timeframe: Immediate. - ACP to use more granular bank-by-bank data for the calculation of risk parameters related to retail lending in TD stress testing. Priority: High. Timeframe: Immediate. - Note: “Immediate” is within one year; “near-term” is 1–3 years; “medium-term” is 3–5 years. ### Additional technical and disclosure notes - The tests relied primarily on BU stress tests and were cross-validated by TD tests; owing to insufficient public disclosure and legal constraints on authorities’ ability to provide supervisory data, the FSAP team could only conduct a partial TD exercise using publicly available information, estimated RWA based on point-in-time PDs, and applied full Basel III with no filters on AFS positions. - The partial TD exercise broadly supported the conclusion that the system was resilient to a wide range of adverse shocks but illustrated sensitivity to assumptions, including probabilities of default and the use of capital measures as defined under CRD IV. - The exercise underscored the importance of improving public disclosure of French financial institutions. _Executive Summary of IMF staff report _cr13185 - Executive Summary_._ ### 16. Solvency stress tests were based on banks’ and ACP’s models: ### 16. Solvency stress tests were based on banks’ and ACP’s models ### Methodology and modelling approach - Banks undertook bottom-up (BU) stress tests using their own models and guidelines developed by the IMF and ACP. - Banks translated key macro-economic variables of the macro-economic scenarios into income, expense, loan loss and capital requirements forecasts. - Loan loss forecasts were disaggregated into Regulatory probability of default (PD) and downturn loss given default (LGD). - Banks added impacts from satellite models (including sovereign risk; counterparty and market risk; funding cost) to assess the global impact on the solvency position. - The ACP conducted top-down (TD) stress tests using supervisory data and two models: 1. Return on asset ratio regression to calculate the stressed capital level. 2. Transition matrices model to calculate the stressed RWA model. - Scope of consolidation for RWA and own funds, profit and loss (P&L), and Balance Sheet (BS) followed the perimeter of the banking group as defined by the CRD. - Future regulatory changes were only captured if they actually came into force during the assessment period (2012 to 2016); phase-in effects were taken into account. ### Sample and scope - BU stress tests included eight banks/banking groups: BNP Paribas, Groupe Crédit Agricole, Société Générale, Groupe Crédit Mutuel, Groupe BPCE, HSBC France, Caisse des Dépôts et Consignations, and La Banque Postale. - TD stress tests bank-by-bank covered the largest potential sample constrained by tool availability: BNP Paribas, Groupe Crédit Agricole, Société Générale, Groupe BPCE, and HSBC France. - The sample of banking groups in the exercise accounted for 97 percent of banking assets. - Stress tests were based on banks’ consolidated exposures, including overseas, and covered only banking operations. - Tests of solvency used bank portfolios as of end-December 2011. - Risk horizon for solvency tests: five years (2012–2016), except: - market and sovereign risks and sensitivity stress tests assessed instantaneous impact as of December 2011. ### Solvency measures and regulatory thresholds - An institution’s solvency was assessed in terms of CET1 plus conservation buffer, and loss absorbency requirement for G-SIBs, annually over the risk horizon. - Ratios were phased in line with Title I of Part Ten of Capital Requirements Regulation (CRR). - Table of minimum capital and buffers (In percent): - Forecast Year: 2012 2013 2014 2015 2016 - Minimum total capital: 8.0 8.0 8.0 8.0 8.0 - Minimum tier 1 capital: 4.0 4.5 5.5 6.0 6.0 - Minimum common equity tier 1: 2.0 3.5 4.0 4.5 4.5 - Conservation Buffer: 0.0 0.0 0.0 0.0 0.625 - Additional Loss Absorbency for G-SIBs: 0.0 0.0 0.0 0.0 0.5 - Footnote: Assumption of 2.0 percent Common Equity Tier 1 requirements in 2019 for BNP Paribas, Groupe Crédit Agricole, Société Générale, and Groupe BPCE. ### Stress test scenarios and shocks - Macro-scenario framework: - Two scenarios: baseline and one adverse scenario. - Baseline scenario followed the February 2012 World Economic Outlook update. - Adverse scenario based on a cumulative deviation from the baseline of 2.1 standard deviation of GDP growth for 2012–13 over the last 10 years. - Adverse scenario combined four adverse shocks (generated by the Mascotte macroeconomic model and calibrated): 1. Reduction in external demand caused by a global recession starting in 2012Q1. 2. Worsening of the European debt crisis: sovereign spreads up by 90 bps for France and 160bps for the euro area; fiscal consolidation to reduce fiscal deficit to 3 percent of GDP by 2013. 3. Worsening in banking funding costs leading to a credit contraction of -0.8 percent in 2012 (in addition to current deleveraging bank plans). 4. Increase in sovereign risk causing repricing of bond holdings in bank portfolios (through haircuts). - In BU tests, this shock affected only non-AAA sovereign bonds in the trading and AFS accounts. - Authorities did not include sovereign risk explicitly in TD stress tests; return on asset is assumed to be stressed globally. - Macroeconomic variables under scenarios (Annual percentage change unless otherwise indicated; Staff estimates based on analysis from Banque de France): - Real GDP growth: - 2011 2012 2013 2014 2015 2016 - Baseline: 1.7 0.5 1.0 1.8 1.9 1.9 - Adverse: 1.7 -1.8 0.2 2.4 2.5 2.3 - Long-term Interest rate: - Baseline: 3.3 3.4 3.4 3.5 3.6 3.7 - Adverse: 3.3 4.4 4.1 3.5 3.7 4.1 - Inflation rate: - Baseline: 2.0 2.0 1.3 1.3 1.5 1.7 - Adverse: 2.0 1.9 1.0 0.6 1.1 1.7 - Unemployment rate: - Baseline: 9.7 9.9 10.1 10.0 9.7 9.3 - Adverse: 9.7 10.2 11.0 11.1 10.8 10.5 - Fiscal balance (percent of GDP): - Baseline: -5.7 -5.0 -4.8 - Adverse: -5.7 -3.6 -3.1 - Housing prices: - Baseline: 6.2 1.2 0.4 0.2 0.9 0.4 - Adverse: 6.2 0.1 -4.4 -1.9 2.0 0.1 - US dollar per EUR: - Baseline: 1.4 1.3 1.3 1.3 1.3 1.3 - Adverse: 1.4 1.3 1.2 1.2 1.2 1.2 ### Assumptions on balance sheets, funding, and single-factor tests - Balance sheets assumed to grow with nominal GDP. - Banks’ funding structure (wholesale, deposits, short and long term, official financing) and hedging strategy were assumed unchanged over the time horizon. - Maturing liabilities assumed replaced by similar ones. - Single-factor tests in BU exercise included: i. Liquidation of non-AAA sovereign bonds in HTM portfolio by country: bonds sold at market values as of December 2011 and applying haircuts listed in Appendix II. ii. Failure of the largest five corporate exposures by name, and the largest five corporate exposures for the five countries where each bank is most exposed. iii. Exchange rate shock (U.S. dollar/euro) of +/- 20 percent. iv. Interest rate shock of 200 bps affecting positions in the banking book including income and valuation effects. v. 25 percent shock to real estate prices. vi. Reverse liquidity test assessing maximum potential loss of wholesale funding, by currency, while still meeting contractual obligations and without access to ECB funding. ### Satellite models: sovereign, funding, counterparty, securitization - Sovereign risk measurement (BU tests): - Term structure of sovereign risk assumed to shift upward in parallel for all countries to which French banks were exposed. - Baseline shock derived from the fiftieth percentile of the historical distribution of annual changes of daily yields. - Adverse shock derived from the ninetieth percentile of the historical distribution of annual changes of daily yields. - Haircuts not applied to AAA rated countries in BU tests; AFS filter taken into account; HTM holdings stressed as sensitivity. - Repricing applied to holdings in AFS, For Valuation Only (FVO), and trading accounts. - Exposures to be stressed included all direct and indirect sovereign exposures: - Net direct exposure = gross exposures (long) net of cash short position of sovereign debt (without derivative hedges such as CDS). - Indirect sovereign exposures included both on and off balance sheet exposures. - Sovereign risk concentrated only during the first year of the adverse scenario; haircuts not applied in subsequent years. - AAA-rated countries (as of December 31, 2011 by at least two main rating agencies) list used by authorities: Australia, Austria, Canada, Denmark, Finland, France, Germany, Luxembourg, Netherlands, Norway, Singapore, Sweden, Switzerland, United Kingdom, and the United States. - Phase-out of AFS regulatory filter in stress-test horizon (Phase-in of Basel 3 regulation for AFS filter in percent): - 2011: 0 - 2012: 0 - 2013: 0 - 2014: 20 - 2015: 40 - 2016: 60 - Impact on AFS reserve and capital described as phased (e.g., by 2016 about 80 percent of the impact of the 2012 sovereign shock taken into account, and phased capital impact: 20% of reserve of 2013 in 2014, 40% in 2015, 60% in 2016). - Bank funding costs: - Stress tests incorporated an increase in funding costs as higher interest payments on corporate deposits and on short- and long-term debt for each year of the risk horizon. - Main drivers of increased cost of funding: higher sovereign risk; evolution of short-term and long-term interest rates (wholesale); rise in banks’ credit spreads; drop in value of sovereign assets used as collateral in funding transactions; deposits (retail) noted as pretty insensitive. - Counterparty risk: - Expected credit value adjustments (CVA) losses associated with counterparty credit risk in the trading book were calculated. - Banks modeled PD and LGD of counterparties (or modeled the CDS-spread) using macro-scenario and market risk shocks, then calculated CVA for outstanding derivatives. - Securitization: - All exposures (traditional and synthetic, re-securitizations, and liquidity lines on securitization transactions) with significant risk transfer (in the meaning of the CRD) were in scope. ### ACP top-down credit loss and income projections - ACP’s TD model focused on corporate sector credit risk and used: - A single equation on net return on asset (ROA) to derive stressed capital level. - A transition matrix model to calculate stressed RWA for corporate claims. - Credit and market risk implicitly captured in the ROA equation. - Stressed RWA for retail was not calculated due to lack of data; BU results were used instead. ### Results, key findings, and vulnerabilities - Overall findings support ACP focus to require banks to build up adequate capital and liquidity buffers. - Aggregate conclusion: banking system would be able to meet regulatory ratios under the scenarios, but pockets of vulnerabilities need addressing. - Solvency stress tests indicate all banks have enough capital to cope with the deterioration in the economic environment while meeting new CRD4 requirements. - Conclusions based on BU stress tests and TD stress tests undertaken for validation. - TD results were more macro-sensitive and showed lower CET1 ratios than banks’ BU results due to differences in models and assumptions. - Timing and drivers: - Adverse scenario has largest impact in 2012 and 2013; implementation of CRD 4 does not start until 2013. - Changes in capital ratios in 2012 and 2013 largely driven by credit risk and credit losses (higher provisions) because adverse scenario increases PD of corporate and retail customers. - Changes in RWA limited by mild deleveraging (credit growth follows nominal GDP, projected to decline by 0.7 percent in 2012) and by use of regulatory through-the-cycle PDs. - Recovery phase during 2014–16 helps smooth the impact of CRD IV introduction. - Authorities estimate largest impact from new regulation will occur in 2015 and 2016: a reduction in CET1 ratio estimated by the authorities of about 43 bps and 82 bps, respectively, above the impact of the adverse scenario. - Concentration and market/sovereign risk: - Sensitivity tests show predominance of credit risk from concentrated exposures to large domestic and/or nonresident borrowers in France, Italy, and the United States; other concentration sources include Belgium, Germany, Spain, Switzerland, Netherlands, and the United Kingdom. - Market risk generally limited; sovereign risks are sizable. - BU stress tests indicate shocks to equity and real estate prices have negligible impact on CET1 ratios. - Non-AAA sovereign debt holdings in AFS and trading accounts remain sizeable, particularly to Italy. - A worsening of the euro crisis would cause cumulated losses of capital of about 5 percent of the initial amount of CET1 capital. - A more extreme shock affecting all sovereign holdings (including France) in all books would impact the initial aggregate CET1 capital by an additional 5 percent. - Such sovereign shocks would be significant for a smaller bank with sizeable HTM holdings of French bonds. - Comparative results: - Aggregate BU results form the basis of presented figures; ACP TD results show a higher decline in CET1 for all banks due to more conservative (higher) and cyclical default probabilities for claims other than retail used by authorities. *Source: IMF staff summary of stress-testing exercise, based on analysis from Banque de France and ACP.* ### 40. The differences among the two methodologies can be summarized as follows: ### _cr13185 - 40. The differences among the two methodologies can be summarized as follows: ### Methodology differences between Bottom-Up (BU) and Top-Down (TD) stress tests - BU stress tests (banks): - Based on banks’ own granular information. - Sovereign risk limited to non-AAA holdings of sovereign bonds (in any book) and non-HTM holdings of any rating category. - Application of the filters on AFS holdings. - Stressed risk-weighted assets estimated using through-the-cycle PDs. - Application of draft transitional provision of Title I of Part Ten of CRR, instead of Basel III. - TD stress tests (ACP): - Based on supervisory data. - Bank capital projections based on an income model (that only implicitly includes funding and sovereign risks). - Application of the filters on AFS holdings. - TTC parameters for the calculation of stressed risk-weighted assets. - Application of draft transitional provision of Title I of Part Ten of CRR, instead of Basel III. - Note: The TTC PDs were, however, more conservative and cyclical, for claims other than retail, than those used by banks. ### III. Liquidity stress tests — design and implementation - Sample and calibration: - BU liquidity stress tests used the same sample of eight banks as for solvency stress tests at end-December 2011. - Run-off rates and haircuts on assets calibrated by type using banks’ historical data during the different crises experienced over the past few years. - Scenario structure: - Two scenarios: with and without banks’ access to ECB facilities. - Methodological approach: - Maturity ladder analysis capturing: - (i) the bank’s liquidity needs derived from outflows (closure/partial closure of unsecured wholesale funding market, impact on secured funding, drawings on credit/liquidity lines, run-off of retail deposits; run-off rates applied to every funding source); - (ii) available standby liquidity from inflows (loans maturing, reverse repos, available credit lines; banks continue to fund customers); - (iii) available liquidity buffers under the two scenarios (with and without ECB facilities). In the no-ECB case, stress assumes haircuts on assets due to massive and sudden sales; in the ECB case, current ECB haircuts are applied. - Unit of measurement: - Survival horizon of institutions (up to two years), assessed through the construction of the residual stressed funding gap after sales of liquid assets. - Implementation assumptions (explicit): - Without banks’ access to ECB facilities. - With banks’ access to ECB facilities, based on the bank’s ability to provide adequate collateral. ### Liquidity stress test results and key statistics - Overall finding: - Despite improvements in bank funding profiles during 2011, vulnerability to liquidity shocks was material. - Without recourse to ECB liquidity: - Two banks would face difficulties meeting liquidity needs from outflows (mostly unsecured wholesale funding from banks and other institutions) with available buffers, standby inflows, and asset sales. - One bank cannot meet contractual obligations at one day to one week. - One bank cannot meet contractual obligations from three months to six months. - A two-notch bank downgrade could impose added stress through collateral and margin calls, with significant effects on some banks. - With ECB support: - All banks would pass the test assuming access to ECB liquidity. - Reverse stress test on maximum potential loss of wholesale funding, by currency, while still meeting contractual obligations: - Three of the banks recurring to central bank liquidity above a 5 percent loss of wholesale funding (in the event of a closure of funding markets). - With ECB support, four banks would be able to address up to a maximum loss of about 15 percent of all wholesale funding. - Survival-period results (number of banks which still meet contractual obligations): - Survival periods (columns as in original table): Up to 1 day; Greater than 1 day up to 1 week; Greater than 1 week up to 1 month; Greater than 1 month up to 2 months; Greater than 2 months up to 3 months; Greater than 3 months up to 6 months; Greater than 6 months up to 1 year; Greater than 1 year up to 2 years. - Number of banks which still meet their contractual obligations without ECB support: 7, 6, 6, 6, 6, 5, 5, 5. - Number of banks which still meet their contractual obligations with ECB support: 7, 7, 7, 7, 7, 7, 7, 7. ### IV. Contagion risk — methodology and findings - Network stress-test approaches (ACP): - Two complementary approaches focused on solvency impacts: an initial shock and a contagion mechanism. - Shock considered: “one of the seven banks fails.” - Direct network stress test: used net exposures and based on Cont et al. (2010). Contagion begins when net exposure losses drive a bank below required regulatory capital (fixed at 7.0 percent of RWA). - Reverse network stress test: used bilateral exposures and the model by Gouriéroux, Héam, and Monfort (2012), with the initial shock defined as the loss value on non-interbanking assets for a given bank that triggers at least another bank’s default; the contagion mechanism analyzed bilateral exposures through Merton’s Value-of-the-Firm. - Network characteristics and exposures: - French banking network moderately concentrated; most interbank exposures within it relatively small. - In net exposures: two banks lend 60 percent of all interbank net flows and one bank receives over 40 percent of French interbank flows; these exposures are relatively small (under 3 percent of total assets). - Contagion test results (using information as of December 2011): - A failure of a single bank would result in a CET1 ratio decline to 8.5 percent from 9.9 percent. - Direct and reverse methods combined: - Over the seven scenarios (failure of a single bank), only three scenarios led to the failure of other banks (capital ratio below 7 percent) due to contagion effects. - For the worst-case initial shock from the first technique, the capital ratio of the French banking system (RWA weighted average) would move from 9.9 percent to 9.1 percent, due to contagion effects. - From the second technique, no bank loss would trigger the default of another bank, nor drive another bank to a prudential ratio below 7 percent. As a reverse stress test, the risk of contagion within the French banking system was null. - Worst-case French banking system capital ratio reported: 8.48 percent, representing a loss of 140 basis points. - External contagion risk: - Larger contagion effects may emerge from exposures to non-French bank counterparts in the interbank market (other European banks, U.S. investment banks, banks from Japan and GCC countries). - The Euro money market topology (including French banks) is highly concentrated with relatively few, highly connected core players and many less connected banks. ### V. Conclusion and policy recommendations - Solvency conclusions: - Solvency stress tests indicate that banks could cope with deterioration in the economic environment while phasing in capital requirements under CRD IV, provided sovereign risks are limited to the impact of the shock on non-AAA holdings in the trading and AFS books. - Conclusions based on BU scenario and sensitivity tests and TD stress tests by authorities and the FSAP team; after reconciliations, both methodologies agreed on the final assessment, though TD results were more macro-sensitive and characterized by lower CET1 ratios than banks’ results. - Liquidity conclusions: - Under no recourse to ECB liquidity, two large French banks appear vulnerable to a liquidity shock calibrated on French historical run-off rates and haircuts. - All banks pass when assuming access to ECB liquidity. - Recommendations: - Strengthen ACP’s TD stress testing approaches and disseminate stress test results: - Enhance methodologies to conduct bank-by-bank TD stress tests and collect necessary data (e.g., for calculation of risk parameters related to retail lending). - Liquidity stress tests should be undertaken frequently. - Both solvency and liquidity results should be made publicly available in the Financial Stability Report. *Source: IMF staff summary of stress test chapter (France), December 2011 baseline and stress test exercises.* ### 0.5 percent, for 2013 1 percent. ### _cr13185 - 0.5 percent, for 2013 1 percent. ### Adverse scenario and key shocks - Adverse: Cumulative deviation from baseline of 2.1 standard deviation of GDP growth for 2012–13, driven by: - (1) reduction of external demand caused by a euro area recession; - (2) a fiscal shock resulting from a temporary rise in sovereign spreads and funding cost; and - (3) worsening in banking funding costs that leads to a credit contraction. - Shocks to sovereign spreads: - Size of parallel shift by country: 90 bps for France and 160bps for Euro area (weighted average by gross debt). - Haircuts applied to all sovereign holdings in all portfolios. ### Sensitivity analysis (bottom-up, top-down, FSAP team) - Bottom-Up by Banks: - Souv: Liquidation of non-AAA sovereign bonds in the HTM portfolio, by country, assuming that bonds are sold at market values as of December 2011; and (ii) assuming that market values deteriorated as per agreed haircuts. - CTP1–6: Failure of largest five corporate exposures by name, and the largest five corporate exposures for the five countries where the bank is most exposed. Exposures comprise lending to the private sector and counterparty risk from derivatives. - FX: A U.S. dollar/euro exchange rate shock of +/- 20 percent. - IR: an interest rate shock of 200 bps affecting positions in the banking book (in order to stress IRBB). This shock will assess the impact of changes in interest rate on solvency due to banks’ duration mismatches (including income and valuation effects). - RE: A shock to real estate prices of -25 percent. - Reverse stress tests of liquidity concentration: The maximum potential loss of wholesale funding, by currency that each bank can suffer while still meeting contractual obligations, without access to ECB funding. - Top-Down by Authorities: - Haircuts applied to AAA sovereign debt holdings in all books. - Top-down by FSAP Team: - n.a. ### Risks and factors assessed - Bottom-Up by Banks: - Credit risk (households and corporates, domestic and foreign exposures). - Sovereign risk for non-AAA government bonds (AfS and TB in scenario and HTM in sensitivity analysis). - Counterparty risk in the banking book. - Funding risk. - Market risk, including equity and exchange rate risks. - Interest rate risk in the banking book. - Top-Down by Authorities: - Credit risk (households and corporates, domestic and foreign exposures). - Counterparty risk in the banking book. - Sovereign risk for AAA government bonds in all books. - Funding risk. - Top-down by FSAP Team: - Credit risk (to the extent of data availability: households and corporates, domestic and foreign exposures). - Sovereign risk for GIIPS, Belgium, and France government bonds (AfS, TB, and HTM). The inclusion of government bonds was limited to publicly available data for all banks. - Counterparty risk in the banking book. - Funding risk. ### Calibration of risk parameters, behavioral adjustments, and regulatory standards - Calibration of risk parameters: - Banks’ models for point in time PDs and income. - RWAs were estimated using through-the-cycle PDs. - Change in valuation due to upward shift of the term structure of sovereign risk. - Sovereign risk shock: ninetieth percentile of historical distribution of changes of daily bonds yields for the adverse scenario, and fiftieth percentile for the baseline, using actual maturities. - Authorities/top-down: Model for income (ROA); Transition matrices model and stressed PDs for RWA; RWAs were estimated using through the cycle PDs; Models for credit losses (loss loan provisions) and income; Quasi-IRB approach for RWA; RWA were estimated using point in time PDs. - FSAP Team: Change in valuation due to upward shift of the term structure of sovereign risk; Sovereign risk shock: ninetieth percentile of historical distribution of changes of daily bonds yields for the adverse scenario, and fiftieth percentile for the baseline, using an assumption of five-year maturity. - Behavioral adjustments: - Deleveraging assumption. - Zero payout under stress. - Regulatory standards applied: - CRD IV (Common Equity Tier 1, Tier 1, Total Capital, conservation buffer, loss absorbency requirement for G-SIBs) for each year of the risk horizon. - Capital ratios phased in line with Title I of Part Ten of CRR. - Regulatory filter for AfS positions. - Basel III referenced for FSAP team (Common Equity Tier 1, Tier 1, Total Capital, conservation buffer, loss absorbency requirement for G-SIBs) and phased in line with Basel III. ### Banking sector: liquidity risk (scope and methodology) - Institutions included: - Eight major banks (BNP Paribas, Groupe Crédit Agricole, Société Générale, Groupe Crédit Mutuel, Groupe BPCE, HSBC France, Caisse des Dépôts et Consignations, and La Banque Postale). - Market share covered: 97 percent of the banking system. - Data and baseline date: - December 2011. - Source: institutions’ own data (including off-balance sheet funding activities of banks). - Scope of consolidation: consolidated banking group. - Methodology: - Cash flow-based liquidity stress test using maturity buckets. - Assumption on withdrawals based on banks’ past crisis experiences, benchmarks from previous EBA exercises, and LCR weights, by source of funding. - Assessment of liquidity buffers. - Haircut on assets if sold. - Risks assessed: - Funding liquidity. - Market liquidity. - Regulatory standard tested: - Ability to respond to withdrawals without having access to ECB facilities. ### Banking sector: financial contagion and spillover risks - Institutions included: - Top-down by Authorities: Seven major banks (BNP Paribas, Groupe Crédit Agricole, Société Générale, Groupe Crédit Mutuel, Groupe BPCE, HSBC France, and La Banque Postale) and sovereigns to which French banks are exposed. - Top-down by FSAP Team: n.a. - Market share: - Authorities’ coverage: 95 percent of the banking system. - FSAP Team coverage: 100 percent of the banking system. - Data and baseline date: - Authorities: December 2011. Source: Large exposures database by the ACP. Scope of consolidation: Consolidated banking group. - FSAP Team: September 2011. BIS data on cross-border consolidated banking exposures. - Methodology: - Authorities: Network analysis model developed by the authorities for French banks. - FSAP Team: Network model for spillovers by Espinosa-Vega and Solé (2010). - Note: - 1/ The results for CDC are excluded from figures and charts included in the document. ### Appendix II: Sample shocks to sovereign yields (selected entries) - Table columns: Tentative Haircut (percent) 1/ Change in Yield (bps) — 50th Pctile 90th Pctile 50th Pctile 90th Pctile - Selected country entries (as listed): - Australia 15.20 83.20 0.76 4.16 - Austria 3.20 93.00 0.16 4.65 - Belgium 26.20 108.90 1.31 5.44 - Canada -16.80 72.80 -0.84 3.64 - Chile 59.70 81.90 2.98 4.09 - China 87.10 126.40 4.35 6.32 - Croatia 72.70 157.70 3.63 7.88 - Czech Republic 2.30 96.00 0.11 4.80 - Denmark -3.40 89.50 -0.17 4.47 - Finland -3.70 80.20 -0.18 4.01 - France 2.10 90.40 0.10 4.52 - Germany -9.80 86.60 -0.49 4.33 - Greece 72.80 869.50 3.64 43.44 - Hungary 16.00 228.00 0.80 11.39 - India 38.60 118.00 1.93 5.90 - Indonesia -141.10 321.00 -7.05 16.04 - Ireland 23.60 540.60 1.18 27.02 - Italy 33.40 127.20 1.67 6.36 - Japan -9.90 46.20 -0.49 2.31 - Korea -46.50 -16.10 -2.32 -0.80 - Lithuania 0.00 0.10 0.00 0.00 - Malaysia 0.00 13.00 0.00 0.65 - Netherlands -9.10 87.90 -0.45 4.39 - New Zealand -22.30 86.10 -1.11 4.30 - Norway 16.70 72.20 0.83 3.61 - Portugal 47.40 752.30 2.37 37.60 - Russia 39.00 106.50 1.95 5.32 - Slovenia 130.50 223.50 6.52 11.17 - Spain 34.00 149.20 1.70 7.46 - Sweden 8.60 81.70 0.43 4.08 - Switzerland -4.80 65.10 -0.24 3.25 - Turkey 3.90 172.00 0.19 8.59 - UK -41.00 64.80 -2.05 3.24 - US -43.91 75.84 -2.19 3.79 - Debt-weighted Averages: - Euro Area 14.87 162.05 0.74 8.10 - All countries -10.77 90.43 -0.54 4.52 - Source note: - Source: Bloomberg - Note: Not all dates available for all countries. - 1/ Tentative haircuts use a durantion approximation formula and a 5-year duration assumption. Banks did the own calculations of haircuts bond by bond; therefore actual haircuts, while similar, might differ from the ones above. ### Appendix III: Risk Assessment Matrix — Selected main threats, likelihood and impact - 1. Prolonged recession in advanced economies - Overall Level of Concern: Medium - Likelihood of Severe Realization of Threat in the Next 1–3 Years: High - Expected Impact on Financial Stability if Threat is Realized: high - Key points: - Signs of deceleration in global data and recent European political developments continue to fuel uncertainty. - Given real linkages with the United States, Germany, and large southern Euro Area countries, and financial linkages with the United Kingdom, France is exposed to both export and financial shocks. - A weaker global environment would weaken confidence and lower domestic consumption and investment. - Bank asset quality (and, consequently, equity value) would be affected. - Bank earnings (and therefore ability to recapitalize using internal resources) would be affected due to lower interest margins and higher provisions on nonperforming loans (NPLs). - Increased financial distress and heightened risk aversion would dampen growth by widening spreads and possibly reducing credit supply, amplifying the recession. - 2. Strong intensification of the Euro Area crisis - Overall Level of Concern: Medium - Likelihood of Severe Realization of Threat in the Next 1–3 Years: High - Expected Impact on Financial Stability if Threat is Realized: high - Key points: - The European environment is highly volatile and uncertainties remain on Europe's ability to resolve its debt crisis. - Any widening of the crisis in Europe could result in very weak growth for the Euro Area. - Continuing uncertainty among high-yield Euro area countries could see sovereign spreads widen further. - French SIFIs’ exposure to high-yield Euro area countries (Italy in particular) is large and could translate into bank losses, via declines in loan quality and lower sovereign bond values. - Further deleveraging by French banks may lower returns if this involves disposing of profitable activities. - 3. France’s sovereign rating downgrade - Overall Level of Concern: High - Likelihood of Severe Realization of Threat in the Next 1–3 Years: Medium - Expected Impact on Financial Stability if Threat is Realized: high - Key points: - France’s sovereign rating faces the risk of further reductions because of its relatively high public debt stock and budget deficit relative to other AAA-rated European countries. - Authorities committed to meeting fiscal targets of France’s Stability Program; additional efforts may be required in 2013, if growth is lower than the current official forecast of 1.7 percent. - As of end-2011, the exposure of the eight largest French banking groups to their sovereign was 2.5 percent of their consolidated assets. - Impact on funding costs of banks and corporates due to contagion could be material if not already factored in by markets. - Possible spillovers from sovereign rating downgrade to banks’ ratings. - The downgrade in credit ratings could affect French banks’ large derivative business and, through higher margin calls, may put pressure on their liquidity. - 4. Closure of wholesale funding to French banks - Overall Level of Concern: Medium - Likelihood of Severe Realization of Threat in the Next 1–3 Years: High - Expected Impact on Financial Stability if Threat is Realized: high - Key points: - Long-term funding improved significantly in the first months of 2012, following the ECB’s three-year LTROs, but has recently become more difficult. Market conditions are still uncertain for the rest of the year. - French banks have significant refinancing needs in 2013–14, and market access may be challenging. - Some French SIFIs’ business models are under pressure if low funding costs are unavailable. - French banks are reliant on significant amounts of wholesale funding. - The domestic interbank market remains partly frozen. French banks, like other European banks, have increased their access to the ECB window. Potential negative second round-effects relate to cross-border interbank exposures and derivatives positions. This could create conditions for a systemic liquidity shock. *Source: _cr13185 - 0.5 percent, for 2013 1 percent.* ### 5. Housing ### 5. Housing ### Staff assessment - Staff assessment: Medium - Staff assessment: Medium-Low ### Drivers of housing prices - Housing prices have been supported until recently by fundamental factors, asset allocation factors (the perception of real estate as a safe haven), and a low interest rate environment. ### Valuation - Staff estimates suggest that France’s housing prices were 10–25 percent overvalued at end-2010. ### Downside risks - Downside risks remain due to high LTV ratios, a worsening of the economic outlook, and a possible unexpected increase in interest rates. ### Risks to banks and households - Risks to banks appear limited as French households have comparatively low levels of debt and lending standards are overall sound notwithstanding some relaxation during the boom years. - However, given the large share of real estate in total households’ net wealth, there are potentially indirect effects via the impact of a real housing price correction on confidence and, thus, on GDP growth (as evidenced during the 2008–09 down cycle). *Source: _cr13185 - 5. Housing* ### 22. The latest WEO update (February 2012) provides the following scenario: ### _cr13185 - 22. The latest WEO update (February 2012) provides the following scenario ### Adverse macro scenario: triggers and overall characterization - Stress scenario triggered by one-year external demand and funding shocks starting in Q1, 2012, impacting mainly 2012–13; subsequent three-year path depends on authorities’ macroeconomic model. - Characterized by volatility in sovereign markets, leading to consolidation of public finances in France that enables a return to baseline. - Shock leads to a maximum cumulated deviation from baseline of 2 standard deviation of GDP growth for 2012–13. - Key triggers: - A reduction in external demand from a global recession starting in Q1, 2012, reducing euro area output by about 4 percent on average relative to the WEO forecast; world output lower than WEO projections by about 2 percent. - Nominal exchange rate depreciation of about 13 percent over 2012–13. - Worsening of the European debt crisis pushing sovereign spreads up to short-term Euro area policy rates by about 160 bps in the Euro area in 2012, motivating fiscal adjustment; sovereign shock represented by ninetieth percentile of historical annual changes of daily sovereign yields (Bloomberg generic five-year government bonds). - Banking funding cost worsening leading to credit contraction and higher corporate funding costs that reduce investment. ### Key macroeconomic and external assumptions (selected exact figures from Baseline and Adverse scenarios) - Euro area real GDP growth: - Baseline 2012: -0.5; 2013: 0.8 - Adverse 2012: -2.9; 2013: -0.5 - World real GDP growth: - Baseline 2012: 3.3; 2013: 3.9 - Adverse 2012: 2.0; 2013: 3.3 - Nominal appreciation (increase in US dollar value of local currency, in percent): - Baseline 2012: -7.8; 2013: -0.1 - Adverse 2012: -6.5; 2013: -6.3 - Petroleum spot price (US dollar, percent change from previous year): - Baseline 2012: 1.6; 2013: -3.1 - Adverse 2012: -13.7; 2013: -1.0 - Nonfuel commodity price (index, percent change from previous year): - Baseline 2012: -10.4; 2013: -1.7 - Adverse 2012: -17.3; 2013: -2.9 - Source of external assumptions: IMF World Economic Outlook Update, January 24, 2012. ### France baseline vs adverse scenario projections (selected exact series) - Real GDP (percent): - Baseline: 2011: 1,7; 2012: 0,5; 2013: 1,0; 2014: 1,8; 2015: 1,9; 2016: 1,9 - Adverse: 2011: 1,7; 2012: -1,9; 2013: 0,0; 2014: 2,4; 2015: 2,6 - Private consumption (percent): - Baseline: 2011: 0,3; 2012: 0,2; 2013: 1,0; 2014: 1,9; 2015: 2,1; 2016: 2,1 - Adverse: 2011: 0,3; 2012: -1,0; 2013: -1,5; 2014: 1,2; 2015: 2,6 - Gross fixed capital formation (percent): - Baseline: 2011: 2,9; 2012: 1,3; 2013: 1,4; 2014: 2,5; 2015: 2,9; 2016: 2,9 - Adverse: 2011: 2,9; 2012: -4,5; 2013: 1,6; 2014: 3,9; 2015: 3,7 - Exports (goods and services, percent): - Baseline: 2011: 5,0; 2012: 2,4; 2013: 2,2; 2014: 3,5; 2015: 3,6; 2016: 3,7 - Adverse: 2011: 5,0; 2012: -0,4; 2013: 1,9; 2014: 5,5; 2015: 5,7 - Imports (goods and services, percent): - Baseline: 2011: 5,0; 2012: 0,2; 2013: 1,9; 2014: 2,5; 2015: 3,2; 2016: 3,4 - Adverse: 2011: 5,0; 2012: -5,2; 2013: 1,3; 2014: 4,4; 2015: 4,9 - Household saving ratio (% of RDI): - Baseline: 2011: 16,9; 2012: 16,4; 2013: 16,4; 2014: 16,3; 2015: 16,1; 2016: 16,1 - Adverse: 2011: 16,9; 2012: 13,9; 2013: 14,0; 2014: 15,9; 2015: 16,3 - Unemployment rate (% of labour force) (e): - Baseline: 2011: 9,7; 2012: 9,9; 2013: 10,1; 2014: 10,0; 2015: 9,7; 2016: 9,3 - Adverse: 2011: 9,7; 2012: 10,2; 2013: 11,1; 2014: 11,2; 2015: 10,9 - GDP deflator: - Baseline: 2011: 1,6; 2012: 1,7; 2013: 1,5; 2014: 1,6; 2015: 1,8; 2016: 2,0 - Adverse: 2011: 1,4; 2012: 1,3; 2013: 0,2; 2014: 0,3; 2015: 0,9 - Public Balance (% of GDP): - Baseline: 2011: -5,7; 2012: -4,9; 2013: -4,7 - Adverse: 2011: -5,7; 2012: -3,6; 2013: -3,1 - Current Account (% of GDP): - Baseline: 2011: -2,3; 2012: -1,9; 2013: -1,5 - Adverse: 2011: -2,4; 2012: -0,7; 2013: -0,8 - Oil price (US$ / barrel): - Baseline: 2011: 111,5; 2012: 105,7; 2013: 102,4; 2014: 98,0; 2015: 94,8; 2016: 92,7 - Adverse: 2011: 111,5; 2012: 89,8; 2013: 88,8; 2014: 93,5; 2015: 93,9; 2016: (not stated) - U.S. dollars per EUR: - Baseline: 2011: 1,39; 2012: 1,28; 2013: 1,28; 2014: 1,28; 2015: 1,27; 2016: 1,26 - Adverse: 2011: 1,41; 2012: 1,31; 2013: 1,21; 2014: 1,21; 2015: 1,2 ### Financial sector stress assumptions: corporate risk, interest rates, and credit - Corporate risk premium increases assumed: - Increase in corporate risk premium of 60 bps in 2012 and 300 bps in 2013 (non-additive to sovereign risk premium). - Interest rates: - Short-run policy interest rates assumed to remain at baseline levels under the stress scenario; only long-term rates shock arises from sovereign risk premium shock. - Domestic credit growth: - Adverse funding conditions in 2012–13 estimated to amount to a 2.8 percent lower growth in domestic credit (relative to baseline, additive to credit-demand effects from lower GDP). - Relation between changes in bank funding costs and stock prices based on recent historical data. ### Satellite models, impairments, and solvency testing - Banks must translate macro variables into income, expense, loan loss and capital requirements disaggregated into Regulatory probability of default (PD) and Downturn LGD. - Impairment parameters: - Based on default rates / default probabilities (pds point in time), loss given default (lgd point in time), and exposures at default (EAD) for domestic and foreign clients, by country. - Risk-weighted assets based on regulatory parameters (through-the-cycle) and adjusted under stress to reflect higher risk. - Equity investments (AFS and fair value through profit and loss) subject to trading-book valuation approach. - All participations subject to IFRS impairment test in baseline and adverse scenarios. - Banks must aggregate satellite model impacts (including sovereign, counterparty and market risk, funding cost) to assess global solvency impact. ### Sovereign risk measurement and treatment in BU and TD tests - Sovereign yields repriced via parallel upward shifts in term structure for all countries to which French banks are exposed. - Baseline shock derived from fiftieth percentile of historical distribution of annual changes of daily yields (Bloomberg generic five-year government bond yields, 2006–11). - Adverse shock derived from ninetieth percentile of the same historical distribution. - When no debt instrument available, haircuts interpolated. - Change in yields used in BU approach to reprice all bonds using cash-flow approach; change in adverse scenario is added to baseline change. - Holdings of government bonds in AFS, FVO, and trading accounts will be repriced; haircuts not applied to AAA rated countries. - For AFS positions, Basel 3 phase-in for AFS filter applied as follows: - 2011: 0%; 2012: 0%; 2013: 0%; 2014: 20%; 2015: 40%; 2016: 60% - Impact on capital: none if no credit default event through 2013; from 2014 onwards phased portion of reserve impacts capital (20% of reserve of 2013 in 2014; 40% of reserve of 2014 in 2015; 60% of reserve of 2015 in 2016). - Exposures to be stressed include all direct and indirect sovereign exposures; net direct position defined as gross exposures net of cash short position of sovereign debt (without derivative hedges). - Direct derivatives positions subject to fair value adjustments using relevant shocks and CVA adjustments; indirect exposures (e.g., CDS) treated similarly. - Haircuts to AFS portfolios applied to adjusted (marked to market) balance sheet values; loans and HTM excluded from haircut calculation but must be stressed via credit risk approach with regulatory PDs and LGDs consistent with scenario. - Sovereign haircuts assumed concentrated only during the first year of the adverse scenario and applied only to that year. ### Bank funding costs and pass-through - Stress incorporates increase in funding costs reflected as higher interest payments on corporate deposits and short- and long-term debt. - Funding-cost evolution modeled via banks’ ALM models, giving particular weight to recent crisis observations; initial increase expected to follow sovereign cost increase closely. - Drivers of increased bank funding costs: - higher sovereign risk; - evolution of short-term and long interest rates (wholesale); - rise in banks’ credit spreads; - drop in value of sovereign assets used as collateral in funding transactions; and - deposits (retail) are relatively insensitive. - Implementation specifics: - For each year of risk horizon, all short-term debt funded at the new funding rate; long-term debt due in each year re-priced at the new rate. - Interest rates on existing floating rate debt reflect expected increases. - Deposit rate adjustments account for very low sight deposit rates and regulated savings products (“livrets réglementés”). - Change in funding costs weighted by levels of long- and short-term debt. - Both cases—with and without pass-through of higher funding costs to customers—will be reported; assumption that due to competition not all increases can be passed through. - Replacement rate of deposit income moves only with evolution of short and long term interest rates (impossible to pass rise in bank’s credit spread). - Banks’ credit spreads evolutions: - Correlated to sovereign credit spreads for France for domestic activities and to relevant subsidiary sovereign spreads for foreign funding. - If no evidence to the contrary, bank credit spread increase applied on a one for one basis (i.e., 100 percent). ### Market risk assumptions in adverse scenario - Market risks incorporated as an instantaneous shock affecting only banks’ first-year results. - Applied following European directives or Basel rules; banks assess market risk using own models and current regulations. - Adverse scenario main market risk assumptions: - Short- and long-term interest rates will not change during the first year, but credit spreads (including bank and corporate debt) will evolve in line with sovereign risk. - An equity shock of about 25 percent to the CAC index. - The U.S. dollar/euro exchange rate changes by +/-6.5 percent, consistent with the WEO adverse scenario assumptions. *Source: _cr13185 (IMF staff content provided in the chapter text).* ### 63. For the computation of the impact from the market risk shocks, the assumption ### _cr13185 - 63. For the computation of the impact from the market risk shocks, the assumption ### Market risk shocks and trading book - Instantaneous shocks (both baseline and adverse) are applied to trading book positions as of March 1, 2012 (paragraph 63). - Different portfolios and books will be stressed using the most appropriate parameters from the set provided (paragraph 63). - For the period 2012 to 2016, net trading income should evolve proportionally to GDP growth (paragraph 64). - Results (losses or gains) of the shocks (sovereign + others trading book portfolio) are used to adjust net trading income forecasts for 2012; no market shock is assumed for 2013 to 2016 in the baseline scenario (paragraph 65). - Parameters may be directional, allowing compensation between gains and losses on different portfolios (paragraph 66). - RWA on market risk (standard and internal models), excluded securitization, should evolve correspondingly to the evolution of trading activities, taking into account balance sheet growth assumptions (paragraph 67). - Banks must derive their own RWA forecast for 2012 to 2016, benchmarked with their QIS submission (paragraph 68). ### Counterparty risk (CVA) - Expected CVA losses associated with counterparty credit risk in the trading book: banks model PD and LGD of counterparties (or model the CDS-spread) using the macro-scenario and market risk shocks, then calculate CVA for outstanding derivatives (paragraph 69). ### Securitization - Scope: all exposures (traditional and synthetic, re-securitizations, liquidity lines on securitization transactions) with significant risk transfer as per CRD (paragraph 70). - The exercise takes into account the CRD III (paragraph 71). - Stress applied to securitization positions (Standard and IRB portfolios) in the different credit quality steps at December 2011; direct increased risk weighted in substitution of the original ones; increase in RWA should reflect macro-scenario and stressed market parameters (paragraph 72). - All positions stressed independently of approach (external rating, Supervisory Formula, Internal Assessment Approach) (paragraph 73). - Exposure currently deducted from capital should be risk weighted at 1250 percent (paragraph 74). - Losses: banks estimate amount of impairment at end of each period taking into account the macro-scenario (paragraph 75). ### Profits projections and P&L assumptions - Profits components (interest income, interest expenses, net fee and commission income, operating expenses) under stress based on satellite model or expert judgment; each profit component estimated separately to the extent possible (paragraph 76). - For end-2011, net profits before tax adjusted for extraordinary income/losses (paragraph 76). - Profit projections include impact of higher interest payments due to higher funding costs during stress (paragraph 76). - Banks shall use FINREP “Consolidated income statement” definitions for forecasting P&L 2012–2016 (paragraph 77). - Trading income 2012–2016 assumed to follow GDP evolution with specific shock in 2012; decline in GDP results in lower trading income (paragraph 78). - Interest income impacts estimated by banks using internal procedures (e.g., ALM tools) respecting methodological note assumptions (paragraph 79). - Specific assumptions on cost of funding in the adverse scenario (paragraph 80). - Roll-over of maturing assets shall not take into account increased credit spread (banks shall not benefit from increase on sovereign or corporate bond yield, nor increase margins on loans) (paragraph 81). - Banks shall not take into account possible debt valuation adjustments (DVA); deterioration of own creditworthiness cannot be booked as P&L profit on net liabilities (paragraph 82). ### Behavioral adjustments (balance sheet, dividends, hedging) - Balance-sheet growth: in principle, balance-sheets grow with nominal GDP; asset allocation remains proportionally identical; no asset disposals/acquisitions except legally binding commitments or approved deleveraging plans; maturing exposures replaced by similar ones unless good reason not to (paragraph 83). - If a firm falls below any regulatory solvency ratios (Table 1), the firm reports actions to respond and transmits additional stress results with those actions taken into account (paragraph 84). - Banks compute two sets of results: (i) constant structure of balance sheet; (ii) taking into account structural plan for implementation of CRR (paragraph 85). - Dividend payout ratio: dividends in stress scenarios only by banks that remain adequately capitalized and after setting aside provisions and reserves (paragraph 86). - Hedging positions expected to be rolled-over with no change in hedging strategy; fair value of hedges must reflect evolution of assets; increased cost of roll-over reflected in P&L (paragraph 87). ### Solvency measures and regulatory treatment - Solvency measured each year of risk horizon in line with transitional provision of Title I of Part Ten of CRR; when CRR offers options, the one in line with Basel III should be taken into account (paragraph 88). - Assessment includes all components of capital (CET1, Tier 1, Total Capital, conservation buffer, loss absorbency requirement for G-SIBs) phased in line with Title I of Part Ten of CRR (paragraph 88). - Calculation of risk weights reflects implementation of new regulatory rules in line with CCR, taking into account regulatory transition for floors (transition from Basel I to Basel II and CRR) (paragraph 88). - Changes in risk weights should reflect at minimum results of the Comprehensive QIS; banks use internal models and may use expert judgment to reconcile business volume changes with new capital charges (paragraph 89). ### Appendix Table 4: France: Solvency Measures Under Stress (In percent) - Forecast Year: Y1 (2012), Y2 (2013), Y3 (2014), Y4 (2015), Y5 (2016) - Minimum Total Capital: 8.0 8.0 8.0 8.0 8.0 - Minimum Tier 1 Capital: 4.0 4.5 5.5 6.0 6.0 - Minimum Common Equity Tier 1: 2.0 3.5 4.0 4.5 4.5 - Conservation Buffer: 0.0 0.0 0.0 0.0 0.625 - Additional Loss Absorbency for G-SIBs: 0.0 0.0 0.0 0.0 0.5 - 1/ Assumption of 2.0 percent Common Equity Tier 1 requirements in 2019 for BNP Paribas, Groupe Crédit Agricole, Société Générale, and BPCE. ### Liquidity stress tests: methodology and unit of measurement - Combined BU and TD rounds; maturity ladder analysis captures: (i) liquidity needs from outflows including unsecured wholesale funding closure, secured funding impacts, extra cash-outflows from drawings and deposit run-off (run-off rates applied); (ii) available standby liquidity from inflows (maturing loans, reverse repos, available credit lines); (iii) available liquidity buffers under two scenarios: with and without banks’ access to ECB facilities (paragraph 90). - TD second round: authorities aggregate counterbalancing measures reported by banks; TD analysis may adjust haircuts or repo roll-over assumptions based on aggregated BU reports (paragraph 93). - Unit of measurement: survival horizon of institutions (up to two years) assessed through residual stressed funding gap after asset sales; implemented under two assumptions: (i) without ECB access; (ii) with ECB access based on ability to provide adequate collateral (paragraphs 94). - Assessment date: end of December (paragraph 95). - Final calibration (closure of wholesale markets, loan decisions, asset haircuts) reported by authorities to FSAP team with results (paragraph 96). ### Sensitivity stress tests (instantaneous shocks as of December 2011) - Tests to be performed BU by banks measuring instantaneous impact on solvency as of December 2011 (paragraph 97). Shocks include: - Liquidation of non-AAA sovereign bonds in HTM portfolio, by country, sold at market values as of December 2011; and scenario where market values deteriorated as per Appendix I, columns 1 and 2. - Failure of largest five corporate exposures by name, and largest five corporate exposures for the five countries where the bank is most exposed (exposures include private sector lending and counterparty risk from derivatives). - Exchange rate shock (U.S. dollar/euro) of +/- 20 percent. - Interest rate shock of 200 bps affecting banking book positions (to stress IRBB), capturing duration mismatches (income and valuation effects). - Shock to real estate prices of negative 25 percent. - Maximum potential loss of wholesale funding by currency that each bank can suffer while still meeting contractual obligations, without access to ECB funding (reverse stress tests of liquidity concentration). - Solvency measures in Table 1 used for solvency sensitivity tests to reflect relative strength among banks (paragraph 98). ### Output, reporting, and metrics - Authorities will: - Document models, exact assumptions and calibration used in TD and BU stress tests and results. - Require from banks details of satellite models and expert judgment used to calibrate tests, including earnings capacity, market, credit, counterparty losses, and change in funding conditions under stress. - Engage with firms to ensure consistency of assumptions and model suitability prior to submission. - Ensure reliability of results by checking against historical experience, other firm stress tests, and TD results. - During the second IMF mission, FSAP may have access to additional details while preserving confidentiality (paragraph 99). - Capital adequacy under stress reported to FSAP using solvency metrics for each year over forecast horizon for aggregate system, including dispersion measures like standard deviation and inter-quartile range (Q3-Q1); reports document satellite models/expert judgment and estimation of important parameters (paragraph 100). - Liquidity stress test outcomes reported on banks’ abilities with and without ECB support; when ECB support included, report specifies whether bank qualified and extent of possible support (e.g., ability to provide adequate collateral); reported for aggregate system including dispersion measures (paragraph 101). ### Appendix Table 5: Sample of Shocks to Sovereign Yields — selected entries (Change in Yield (bps) and Tentative Haircut (percent)) - Austria: 50th Pctile 2.95 90th Pctile 2.69 50th Pctile haircut 0.15 90th Pctile haircut 4.63 - Belgium: 50th Pctile 26.50 90th Pctile 107.68 50th Pctile haircut 1.32 90th Pctile haircut 5.38 - Czech Rep.: 50th Pctile -30.95 90th Pctile 55.68 50th Pctile haircut -1.55 90th Pctile haircut 2.78 - France: 50th Pctile 1.40 90th Pctile 90.37 50th Pctile haircut 0.07 90th Pctile haircut 4.52 - Germany: 50th Pctile -9.95 90th Pctile 86.57 50th Pctile haircut -0.50 90th Pctile haircut 4.33 - Greece: 50th Pctile 72.75 90th Pctile 869.32 50th Pctile haircut 3.64 90th Pctile haircut 43.46 - Italy: 50th Pctile 33.35 90th Pctile 126.37 50th Pctile haircut 1.67 90th Pctile haircut 6.32 - Portugal: 50th Pctile 47.70 90th Pctile 723.56 50th Pctile haircut 2.38 90th Pctile haircut 36.17 - Spain: 50th Pctile 34.00 90th Pctile 146.85 50th Pctile haircut 1.70 90th Pctile haircut 7.34 - United States: 50th Pctile -44.05 90th Pctile 75.80 50th Pctile haircut -2.20 90th Pctile haircut 3.79 - Simple Averages: Euro Area 50th Pctile 29.93 90th Pctile 241.19 50th haircut 1.50 90th haircut 12.06 - Simple Averages: All Countries 10.63 170.65 0.53 8.53 - Weighted Averages (by Gross Debt in U.S. dollar): Euro Countries 13.37 159.98 0.69 8.00; All Countries -20.42 75.06 -1.02 3.75 - Source: Bloomberg (note in original: banks did their own bond-by-bond haircuts; tentative haircut uses duration approx. with five-year duration assumption). ### Appendix Table 6: Key scenario and solvency parameter excerpts (selected numbers) - Baseline / Adverse columns present values across 2011–2016 and 2012–2016 respectively. Selected scenario macro variables: - Real GDP growth (Baseline): 1,7 0,5 1 1,8 1,9 1,9 - Real GDP growth (Adverse): -1,8 0,2 2,4 2,5 2,3 - Nominal GDP growth (Baseline): 3,3 3,3 2,5 3,6 3,9 3,9 - Nominal GDP growth (Adverse): -0,1 0,9 2,9 3,7 4,3 - Unemployment rate (Baseline): 9,7 9,9 10,1 9,8 9,4 9,1 - Unemployment rate (Adverse): 10,2 11 11,1 10,8 10,5 - Nominal short term interest rate (Baseline): 1,4 0,9 0,9 1,3 1,9 2,5 - Nominal long term interest rate (Baseline): 3,3 3,4 3,4 3,5 3,6 3,7 - Inflation rate (household's consumption deflator) (Baseline): 2 2 1,3 1,3 1,5 1,7; (Adverse): 1,9 1 0,6 1,1 1,7 - GDP deflator (Baseline): 1,6 2,8 1,5 1,8 2 2; (Adverse): 1,7 0,7 0,5 1,2 2 - Nominal exchange rate (U.S. dollar in euros) (Baseline): 1,39 1,28 1,28 1,28 1,27 1,26; (Adverse): 1,30 1,22 1,20 1,20 1,21 - Solvency minimums (as reported in table excerpts): - Total Capital: 8 8 8 8 8 8 (baseline columns) - Tier 1: 4 4 4,5 5,5 6 6 (baseline columns) - Common / Core Tier 1: 2 2 3,5 4 4,5 4,5 (baseline columns) - Conservation Buffer: 0 0 0 0 0 0,625 (baseline columns) - G-SIB add-on: 0 0 0 0 0 0,5 (baseline columns) ### Liquidity stress test reporting templates (selected items) - Loss of funding buckets: Up to 1 day; Greater than 1 day up to 1 week; Greater than 1 week up to 1 month; Greater than 1 month up to 2 months; Greater than 2 months up to 3 months; Greater than 3 months up to 6 months; Greater than 6 months up to 1 year; Greater than 1 year up to 2 years. - Aggregated cumulative withdrawal/losses to be reported for: retail deposits; unsecured wholesale funding; secured wholesale funding. - Liquidity gap, aggregated cumulated net funding gap, aggregated cum. net funding gap after counterbalancing capacity without CB support, and with CB support. - Survival period and counts: Number of banks which still meet contractual obligations without ECB support; with ECB support. - Survival period summary statistics: Mean; Interquartile range (Q3-Q1); Standard deviation; Estimated survival period (in days) without ECB support; with ECB support. - Sensitivity stress test: Estimated maximum potential loss of wholesale funding, by currency, while still meeting contractual obligations during at least 1 month (in %); reported for EUR, U.S. dollar, Other (Mean; Interquartile range (Q3-Q1); Standard deviation). *Source: IMF staff note as provided in the content unit.* ### 1. Capital is expected to change for the impact, due to the capitalization of profit or ### _cr13185 - 1. Capital is expected to change for the impact, due to the capitalization of profit or ### Capital assumptions and rules - Capital is expected to change for the impact, due to: - the capitalization of profit or loss after tax, - the phase-in deductions, - the grandfathering of capital instruments non-eligible to Title II of Part Two of CRR, - and/or for the amortization of Tier 2 instruments in the last five years following the current regulatory regime. - Banks must detail other potential changes in capital amounts in the “CRD3_details” and “CRD4_details” of the exercise templates. - Substitutions of capital instruments (step-up clause) by issuances that have the same quality (Core, Tier 1, Tier 2, Tier 3) and amount are allowed within the time horizon, but: - the correct estimation of the cost of capital of the new instruments, reflecting the scenarios, must be reflected in the P&L. - Capital actions and issuances already launched and with funds transferred to institutions in the sample by April 30, 2012 are considered in the exercise. - Capital actions and issuances between January 1 and April 30, 2012 will only be considered if they are publicly announced before December 2011 and fully committed by April 30, 2012. - These items should also be reported in “memorandum items” in the “CAP” template. - The exercise is based on regulatory regime at the date of the stress taking into account transitional provision (see Title I of Part Ten of CRR). - No government support measures are assumed. - Tax regimes are treated like regulatory changes as they are at present moment (December 2011) with changes only if agreed by law and definitely coming in. Deferred tax credits, where applicable, may be recognized. *Defaulted assets flow (A-IRB portfolios and Standardized)* ### Method for estimating new defaulted asset flows - New defaulted assets (flows) are estimated by multiplying expected default rates at the end of the year (PD’pit) by the EAD at the beginning of each year (EADt), gross of funded credit risk mitigation factors. - Formula and timing examples: - At the end of year 2012: - Default Flows (2012) = Exp (2011) * PD’pit (2011) - Exp (2011) = EAD (2011), gross of funded CRM - PD pit (2011) = default rate expected at the end of 2011 for 2012 before the application of the scenario in 2012 - PD’pit (2011) = default rate expected at the end of 2011 for 2012 after the application of the scenario in 2012 - At the end of year 2013: - Default Flows (2013) = Exp (2012) * PD’pit (2012) - Exp (2012) = Exp (2011) – Default Flows (2012) - PD’pit (2012) = default rate expected at the end of 2012 for 2013 after the application of the scenario in 2013 - Definitions and constraints: - PD’pit = individual/asset class probability of default after the application of the stress. - EAD at the beginning of each year should be equal to the exposure value, as defined by the CRD, of the non-defaulted assets, gross of funded CRM and after adjustment for the balance-sheet growth assumption. - Banks should explicitly consider the possible impact of decreases in fair value of credit mitigants and recent events and trends on loan portfolios when estimating PD’pit. - Stock of defaulted assets at the end of each forecast year t (2012 to 2016) = - amount of defaulted assets at end of previous year (t-1) + expected defaulted flows in year t. ### Allocation across rating classes (A)IRB portfolios - Default flows to different rating asset classes should be computed at the end of each year and assigned in proportion to the mean EAD-weighted PD’pit of the class. - Example (preserves source numbers): - If PD’pit (t) after scenario = 10 percent, Exp (t) = 100, portfolio has: - CL1: PD’pit (t) = 7,50% and Exp (t) = 80 - CL2: PD’pit (t) = 20,00% and Exp (t) = 20 - Flow of defaulted assets at end of year t+1 = 10 - Assigned: 6 to CL1 and 4 to CL2 - Resulting portfolio average PD’pit before next scenario = 9,72% and EAD = 90 - Any deviation from this allocation approach must be clearly justified; distribution across rating classes should reflect recent historical rating migrations. ### Standardized portfolios - Banks may use internal models not recognized for capital computation if accepted by the ACP. - If no appropriate internal models, banks are expected to approximate PDpit (before scenarios) using last observation(s) at end December 2011 or the average of the last three years. - For 2012-1016 forecast period banks must estimate evolution of PDpit consistently with macro-scenarios (baseline and adverse) and historical data. - (A)IRB banks are encouraged to extend (A)IRB portfolio PD forecasts to SA and (F)IRB portfolios where appropriate. *Loan losses — Stressed PDs (PIT) and LGDs (PIT)* ### Impairment estimation principles - Impairment flows estimated on both defaulted and non-defaulted assets by applying expected loss impairment rates (LGDpit) to exposures. - New defaulted assets computed by applying stressed default rates (PDspit) on initial EAD (gross of funded CRM) of standardized and (A)IRB portfolios. - LGDpit used for impairments should usually differ from LGD downturn parameter used for RWA calculation for (A)IRB portfolios. - On defaulted (A)IRB assets the best estimate of LGD = LGDpit. - Write-off and positive assumptions regarding increasing recovery flows on defaulted assets are not allowed. - Impairment flows on defaulted assets in year t = impairments on new defaulted assets in year t + increase in impairments of existing defaulted assets in year t-1. - Stock of impairments on non-defaulted assets at end of each year should be recomputed to reflect potential increases in expected losses. ### (A)IRB portfolios: loss computation - For new defaulted assets, (A)IRB banks should make impairments equal to the best estimate of LGD, reflecting analytical evaluation of single defaulted exposures and recent workout trends; long term average downturn LGD used as benchmark where appropriate. ### Illustrative impairment flow formulas (preserve exact text) - At the end of 2012: - Specific prov (2012) = Default Flows (2012) * LGD’pit (2011) - LGD’pit (2011) = loss impairment rate expected at the end of 2011 for 2012 after application of the scenario in 2012 - At the end of 2013: - Specific prov (2013) = Default Flows (2013) * LGD’pit (2012) - LGD’pit (2012) = loss impairment rate expected at the end of 2012 for 2013 after application of the scenario in 2013 - Impairment flows on old defaulted assets: - End of 2012: [Def Stock (2011) * LGD’pit (2012)] – Stock Specific Prov (2011) - Def Stock (2011) = stock of defaulted assets at beginning of 2012 gross of impairments - Stock Specific Prov (2011) = Def Stock (2011) * LGDpit (2011) - End of 2013: [Def Stock (2012) * LGDpit (2013)] – Stock Specific Prov (2012) - Def Stock (2012) = Def Stock (2011) + Default Flow (2012) - Stock Specific Prov (2012) = Stock Specific Prov (2011) + Specific Prov (2012) ### (F)IRB and standardized portfolios - (A)IRB banks are encouraged to extend forecasted evolution of best estimate LGDpit to SA and (F)IRB portfolios where appropriate. - If no internal models for LGDpit, approximate LGDpit before scenarios via last observation of loss rates (2011 yearly impairment flow on new defaulted assets/total new defaulted assets in 2011). - For 2012-1016 forecast period, banks must estimate LGDpit evolution consistently with macro-scenarios (baseline and adverse) and historical data. - Macro-economic scenarios must be reflected in fair value of credit risk mitigants (“CRM,” i.e., financial collateral). *RWA / Credit risk without securitization positions* ### Standardized portfolios: RWA floor and rules - RWA for standardized portfolios should be calculated based on scenarios assuming rating migration as appropriate. - For each forecast year t of the stress test, the RWA at the end of the previous year (year-end “t-1”) should be considered as a floor (e.g., RWA as at end 2011 is a floor for 2012, RWA as at end 2012 is a floor for 2013, etc.). - No rolling out in the application of the (A) IRB models is allowed in the time horizon. ### (A)IRB portfolios: RWA forecasts and floors - RWA forecasts over the stress-test horizon (2012 to 2016) must reflect estimated yearly defaulted/impairment flows and application of new regulatory parameters after stress (new PDs, new LGDs) as estimated by stress test models in use (CRD prescription for internal model authorization). - RWA on (A)IRB portfolio for year t are subject to a minimum floor equal to RWA at end of previous year (t-1). ### Stressed PDs and LGDs; EAD growth assumptions - Adequate stress testing methodologies are required for authorization to use internal rating systems for supervisory capital purposes. - Stress tests include sensitivity analyses and scenario analyses (simultaneous variation in multiple risk factors). - Banks shall use stress testing methodologies to simulate impact on credit capital requirements (due to evolution of regulatory PDs and LGDs) from macro-economic scenarios (baseline and adverse). - For consistency with constrained balance sheet growth (balance-sheets assumed to grow in line with nominal GDP) and identical asset allocation it is assumed performing EADs will grow in line with nominal GDP, keeping same risk profile: - If nominal GDP increases in year t, new originations at beginning of year t should be assigned to rating classes in proportion to allocation of performing EAD at end of previous year. - If nominal GDP decreases in year t, EAD assigned to each rating class will be reduced in same proportion so beginning-of-year performing EAD reflects structure at end of previous year. - Maturing assets each year are assumed to be immediately replaced in the same year by assets with same risk and maturity profile. - Credit capital requirement estimation for 2012 to 2016 must reflect potential transition of exposures between rating classes by remapping each year of individual PDs after scenario application to appropriate rating classes. ### Treatment of IRB excess/shortfall and RWA on defaulted assets - For simplicity and consistency impairments on new defaulted assets = best estimate of LGD. - Difference between LGD downturn and best estimate of LGD, when downturn > best estimate, will be computed as RWA. - Excess/shortfall on old defaulted assets should be changed according to expected evolution in impairments over exercise horizon. - IRB excess/shortfall examples (preserve exact formulas): - End of 2012: Def stock (2011) * [Best estimate LGD (2011) – Best estimate LGD (2012)] = + excess or - shortfall - Def stock (2011) = stock of defaulted exposures at beginning of 2011, gross of impairments - Best estimate LGD (2011) = best estimate of LGD at end of 2011 before application of scenario - Best estimate LGD (2012) = best estimate of LGD at end of 2012 after application of scenario in 2012 - End of 2013: Def stock (2011) * [Best estimate LGD (2012) – Best estimate LGD (2013)] = + excess or - shortfall - Definitions as above for Best estimate LGD (2012) and Best estimate LGD (2013) *Italic source attribution line* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2013/_cr13185.pdf_