## 1canea2025008-source-pdf ## Source details **Canonical URL:** [1canea2025008-source-pdf](https://www.imf.org/-/media/files/publications/cr/2025/english/1canea2025008-source-pdf.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2025/english/1canea2025008-source-pdf.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2025/english/1canea2025008-source-pdf.pdf.json) --- ### Executive summary and scenario overview - Time horizon of FSAP adverse scenario: 2025–27. - Adverse scenario features: - Deepening geoeconomic fragmentation, protectionism, increased cross-border restrictions. - Disruption of global production chains, temporary supply shortages, initial rise in inflation. - Central banks pausing or reversing interest rate cuts; elevated interest rates in initial years. - Falling oil prices as growth slows; market valuation corrections. - Sharp drop in residential real estate prices; equity price corrections; exchange rate depreciation. - Macroeconomic outcome (selected): - Real GDP: cumulative decline of 4.7 percent over the initial two years of the scenario (Executive Summary). - Alternative adverse description in insurance chapter: Real GDP cumulative decline of 4.8 percent over the initial two years (2025: −2.4 percent; 2026: −2.4 percent) and described as a 2.7 standard deviation shock. ### Financial system resilience — overall assessment - System-wide findings: - Financial system remained stable amid sizable swings in output, inflation, and interest rates. - Credit performance of deposit-taking institutions (DTIs) is robust; capital and liquidity buffers exceed regulatory minima. - Nonperforming loans (NPLs) low despite recent uptick; NPL ratio: 0.7 percent (end-2024). - Nonbank financial institutions (NBFIs) — including insurance, pensions, and investment fund sectors — have weathered shocks well. - Structural household vulnerabilities: high debt levels and rising debt-servicing costs compared to G7 peers. ### Household vulnerabilities and mortgage risk - Household indebtedness and dynamics: - Canada’s household debt: exceeding 100 percent of GDP and 170 percent of disposable income as of 2024 (insurance chapter). - Residential mortgages: constitute 74 percent of total credit liabilities of households (2024: Q4). - Mortgage portfolio composition: 80 percent fixed-rate loans; 20 percent variable-rate loans; 75 percent of the variable-rate loans are Variable Rate Mortgage with Fixed Payments (VRMFP). - VRMFP in negative amortization: declined from 31 percent in 2023 to about 12 percent as of September 2024 (representing 2 percent of all mortgages). - As of early 2025, approximately 60 percent of outstanding mortgages will renew by 2026 and around half of them could be subject to a higher interest rate at renewal. - Debt service ratio (DSR) projections: - Observed DSRs: - Overall household DSR peaked at 15.1 percent in 2023: Q4; moderated to 14.7 percent in 2024: Q3. - Mortgage DSR: record 8.2 percent in 2023: Q2; 8.0 percent in 2024: Q3. - Projection paths for mortgage DSR: - Baseline: 8.1 percent in 2024: Q4 → 6.7 percent in 2027Q4. - Adverse: 8.1 percent in 2024: Q4 → peaks at 10.3 percent in 2026: Q4 → 8.6 percent by 2027: Q4. - Mortgage Probability of Default (PD) projections (adverse scenario): - Uninsured loans PD by end-2027: 0.9 percent. - Insured loans PD by end-2027: 1.4 percent. - Key drivers of mortgage defaults: - Unemployment: primary driver for uninsured loans. - Wage growth: primary driver for insured loans. - Methodology notes: - Two-stage approach: structural loan-level DSR simulation using RESL data and a Panel Bayesian Model Averaging (BMA) econometric framework for PDs. - Simulations initialized to end-2024 conditions; 1,000 stochastic simulations per period per scenario. - Projections do not incorporate new mortgage originations after late 2023 (RESL dataset cutoff). ### Corporate vulnerabilities and PDs - Corporate PD projections: - Corporate PDs: 0.5 percent in 2024: Q4 → 1.3 percent by end-2027 (adverse scenario). - Sectoral and firm-size heterogeneity: - Large non-financial corporations: generally stable on leverage and liquidity. - Vulnerabilities concentrated among small and mid-sized businesses, especially consumer-facing sectors (food services, accommodation, retail). - Sectors with pronounced projected PD increases: mining, quarrying, oil and gas, construction, real estate. - Firm-level findings: - Liquidity buffers (cash) materially lower default probabilities; interaction between profitability and cash buffers highly significant. - Firm-level regressions show liquidity constraints as a key driver of default; lower profitability increases the importance of cash buffers. ### Banking sector — solvency stress test results - Sample and scope: - Seven systemic DTIs (six D-SIBs plus Québec’s D-SIFI); represent over 90 percent of DTIs’ assets; cutoff date: end-October 2024 / end-2024. - Aggregate CET1 trajectory (selected): - Starting CET1 in 2024: 13.7 percent (aggregate for seven systemic DTIs). - Baseline projection by 2027: 15.7 percent. - Adverse scenario: average CET1 declines by 2.3 percentage points at the trough. - Hurdles and individual outcomes: - CET1 minimum requirement: 4.5 percent. - Dividend-affecting CET1 threshold: 8 percent. - All seven systemic DTIs remain above 4.5 percent and above the 8 percent threshold under the adverse scenario. - Main drivers of CET1 decline: - Rise in loan losses and increase in risk-weighted assets (RWAs); to a lesser extent decreases in Net Interest Income (NII) and non-interest income and increases in non-interest expense. - Sensitivity analyses (selected outcomes): - Market risk sensitivity: average CET1 falls to 10.4 percent in 2026; all institutions remain above 8 percent. - Higher PDs in manufacturing and commodity sectors: additional CET1 drop of 40 basis points relative to adverse scenario. - Conservative PDs across all portfolios: additional CET1 drop of 1.7 percentage points relative to adverse scenario, reaching 9.8 percent in 2027. ### Banking sector — liquidity stress tests - LCR-based tests (30-day horizon): - Aggregate LCR (October 2024): 136 percent (aggregate for 7 systemic DTIs reported elsewhere as 131 percent in CAD and 120 percent in USD). - Aggregate findings: - Aggregate system liquid and broadly resilient to sizable funding outflows over 30 days. - Under all individual retail scenarios, aggregate LCR remains above 100 percent. - In wholesale and combined scenarios, aggregate LCR falls somewhat below 100 percent. - Individual outcomes: - Three systemic DTIs resilient under all wholesale scenarios. - Four systemic DTIs fall below 100 percent in some wholesale scenarios (not by large amounts). - Cash Flow (CF) analysis (one week to one year, emphasis on three months): - CF indicators: cumulative net funding gap and counterbalancing capacity. - Scenario set: 20 scenarios from mild to aggressive. - Three-month horizon results: - On average, the 7 systemic DTIs maintain liquidity surpluses under most scenarios. - Shortfalls occur only in the three most aggressive scenarios (scenarios 18–20). - Under the most aggressive scenario, aggregate shortfalls amount to 237 billion CAD, representing 12 percent of initial counterbalancing capacity. - Institution-level heterogeneity: one institution begins to experience shortfalls by scenario 12; two institutions maintain surpluses under all scenarios. - LCR and CF assumptions: - Haircut and run-off permutations: 3 market scenarios × 4 run-off rate scenarios = 12 LCR scenarios. - Run-off principle: wholesale run-offs > retail run-offs; insured funding has lower run-off rates. ### Insurance sector — solvency and liquidity analysis - Market size and structure: - Canada ranks ninth globally by written insurance premiums. - Life insurer assets (2023): CAD 1,749 billion. - 56 percent general funds; 44 percent segregated funds. - Life insurance concentration: three largest insurers ~85 percent of market. - Solvency ratios (median federally chartered insurers, 2023): - Life insurers: 172 percent (median). - Non-life (P&C) insurers: 398 percent (median). - Top-down solvency stress test (3-year horizon): - Adverse scenario shocks include equity, interest rates, credit spreads, CRE, FX movements. - Life insurers (median firm) under adverse scenario: - Solvency ratio path: 136 → 168 percent (Y1) → 142 percent (Y2) → 121 percent (Y3) — cumulative decline of 15 percentage points. - Median available capital: increases by 15 percent in year 1, then declines by 18 percent in year 2 and by 9 percent in year 3 under adverse scenario. - None of the life insurers falls below regulatory thresholds; many remain above or close to internal operational capital targets. - P&C insurers (median firm) under adverse scenario: - Solvency ratio path: 259 → 222 percent (Y1) → 218 percent (Y2) → 237 percent (Y3). - Investment and liability dynamics less sensitive than life insurers; median available capital declines then recovers in year 3. - Sensitivity analysis: - Instantaneous 40 percent equity decline: solvency ratios decline by 19 percent (life) and 23 percent (P&C). - CAD appreciation +10 percent: median life insurer solvency ratio declines by <1 percentage point; median P&C insurer declines by 2 percentage points. ### Mortgage insurers (Box 1) — top-down risk analysis - Starting positions: - Pre-stress solvency ratio (weighted average): 183 percent. - LTV distribution: 63 percent insured volume LTV < 70 percent; 38 percent LTV < 50 percent. - Adverse scenario outcomes (selected): - Incurred claims: CAD 290m in 2025; CAD 520m in 2026; CAD 450m in 2026 (source duplicates year). - Combined ratio peaks at 37 percent in 2026 (25 percentage points higher than baseline). - Investment valuation: decline of 7 percent in year 1. - Available capital: increases in baseline and adverse but ~11 percent lower in adverse in year 2 relative to baseline. - Solvency coverage: increases to 193 percent in year 2 under adverse (21 percentage points lower than baseline). ### Pension sector — solvency and liquidity analyses - Sector size and composition: - Assets of trusteed pension funds (2023): CAD 2.2 trillion (75 percent of GDP). - Four-fifths of assets held by public sector funds. - Over 16,000 plans; “Maple Eight” are large DB public plans. - Top-down solvency stress test (40 DB and hybrid schemes; three-year horizon): - Starting funding: median funding ratio around 120 percent in 2023/24 for a representative sample. - Adverse scenario (median large public plan and private median plan): - Year 1: mixed effects; some public plans see funding ratio increases; median private plan increases by 4 percentage points. - Year 2: funding ratios fall by 13 percentage points (median large public plan) and by 16 percentage points (median private plan). - Year 3: funding ratios stabilize and increase slightly. - By end of year 3: a bit more than a quarter of plans would experience a funding deficit (no immediate refunding requirement). - Funding ratio levels (selected): - Baseline, large public sector plan: 128 → 140 percent by year 3. - Baseline, median private sector plan: 118 → 132 percent by year 3. - Adverse, median large public sector plan: reach 112 percent by year 3. - Adverse, median private sector plan: reach 107 percent by year 3. - Bottom-up liquidity (margin-call) test — sample and results: - Sample: 19 pension plans (aggregate assets ~ CAD 1.83 trillion, ~70 percent of sector) and 3 large life insurers (assets ~ CAD 1.70 trillion, ~85 percent of life sector). - Tail shock calibration: short-term rates +150 bps; long-term rates +50 bps; CAD depreciation: −2.9 percent vs USD, −4.8 percent vs EUR. - Aggregate margin calls: CAD 32 billion total (pensions: CAD 26.4 billion; life insurers: CAD 5.5 billion). - Of pension margin calls, CAD 24.0 billion due within t+1. - Margin calls as share of combined balance sheet assets: around 1 percent. - Timing: ~81 percent of margin and collateral calls due within one day. - Sources of liquidity: - Pension plans would source CAD 39 billion to meet margin calls: 55 percent from financing transactions (repos and credit lines); 31 percent from highly liquid assets; remainder from cash and deposits. - Life insurers: rely mostly on liquid assets (91 percent of sourced liquidity). - Settlement in kind: 53 percent of aggregated margin and collateral calls can be settled in kind; life insurers able to settle almost all margin calls in kind. - Typical valuation impact on liquid assets: decline by 2.2 percent on average; range for half the sample: −1.7 to −4.2 percent. ### Interconnectedness and contagion analysis - Scope and method: - Evaluated funding and credit shock propagation among six D-SIBs and across sectors and borders using Espinosa-Vega Sole (EVS) network model. - Cross-sectoral and cross-border exposures assessed via “from-whom-to-whom” statistics from Statistics Canada. - Key parameters used in contagion modelling: - λ = 0.4 and 1.0 (LGD on exposures). - ρ = 0.35 (share of lost funding that is non-replaceable). - δ = 0.5 and 1.0 (discount on asset sales). - For cross-sectoral and cross-border estimates: 40 percent LGD and 50 percent discount applied. - Main findings: - Average losses due to credit and funding shocks could reach approximately 5 percent of capital across various banks. - Significant interbank exposures exist, but spillover risks from bank failures are mitigated by banks’ capital buffers. - NFCs and the external sector: highest contagion levels but relatively low vulnerability. - Households, banks, and NBFIs: less contagious but more vulnerable due to high household debt and market sensitivity. - Cross-border risks: most domestic sectors appear very vulnerable to external shocks (except investment funds, insurance companies, OFI and NFC). - Implication: Vigilant monitoring and analysis of cross-sectoral and cross-border transmission channels required. ### Key quantitative highlights (selected, verbatim) - Scenario horizon: 2025–27. - Real GDP cumulative decline: 4.7 percent over initial two years. - Number of systemic DTIs stress-tested: seven. - Average CET1 decline at trough: 2.3 percentage points. - CET1 minimum requirement: 4.5 percent. - Dividend-affecting CET1 threshold: 8 percent. - Mortgage DSR: 8.1 percent in 2024: Q4 → 10.3 percent by end-2026 → 8.6 percent thereafter. - Mortgage PDs by end-2027 under adverse scenario: 0.9 percent (uninsured), 1.4 percent (insured). - Corporate PDs: 0.5 percent in 2024: Q4 → 1.3 percent by end-2027. - Margin calls in severe liquidity scenario: CAD 32 billion total; 26 billion for pension plans. - Margin calls share of combined balance sheet assets: around 1 percent. - Total assets of financial institutions (2024): 756 percent of GDP. - Nonbank financial intermediation share: about 60 percent of financial system assets. - Six largest banks account for 94 percent of the sector's assets; two largest represent 47 percent. - Loans constitute 50 percent of total assets for the seven systemic DTIs (end-2024); residential mortgage loans make up 40 percent of the total loan portfolio. - ROA for the 6-DSIBs: around 0.7 percent. - Average ROE for Canadian banks: 12 percent as of 2024. - Average CET1 ratio for the 6 D-SIBs: 13.3 percent as of end-2024. - Average LCR of the 6 largest banks: 125 percent. ### Policy implications and recommendations (selected, verbatim where provided) - Maintain vigilant monitoring and analysis of cross-sectoral and cross-border transmission channels given high interconnectedness. - Continue monitoring household vulnerabilities (high debt, rising debt-servicing costs), with focus on renters and lower-income households. - Monitor small and mid-sized enterprises in consumer-facing sectors for increased insolvency risk. - Ensure banks retain adequate capital and liquidity buffers given potential rises in loan losses and RWAs under adverse macro scenarios. - Monitor liquidity risks in insurers and pension plans, including preparedness for margin calls and the use of contractual in-kind settlement mechanisms. - Strengthen financial-sector supervisory and data capabilities (selected action items): - Establish regular coordination meetings among BOC, OSFI, AMF to discuss stress testing methodologies, results, and implications. (Responsible: BOC, OSFI, AMF; Timing: ST) - Continue developing structural models for households and corporates supported by continuous investment in rich datasets. (Responsible: BOC, OSFI, AMF; Timing: ST) - Start building a credit register for corporate, commercial and consumption loans and continue improving Real Estate Secured Lending (RESL) data quality and coverage. (Responsible: OSFI, AMF; Timing: ST) - Collect long historical time-series data on default rates across sectors and asset classes by country of exposure. (Responsible: OSFI, AMF; Timing: ST) - Calculate provisions in DTIs top-down stress test exercises in accordance with the International Financial Reporting Standards (IFRS) 9 framework. (Responsible: BOC, OSFI, AMF; Timing: MT) - Develop a top-down approach for assessing systemic DTIs’ market risk that evaluates sudden market shocks while factoring in hedging strategies. (Responsible: BOC, OSFI, AMF; Timing: ST) - Introduce macroprudential stress testing in the insurance sector. (Responsible: OSFI, AMF in cooperation with BOC; Timing: MT) - Expand supervisory reporting for large pension plans (investments, derivatives, liquidity, leverage) and roll out LCR monitoring to all large plans. (Responsible: OSFI, FSRA, FA, PA, BOC; Timing: ST) - Consider publishing regularly top-down stress test results on an aggregate basis in the Financial Stability Report and/or Staff Analytical Notes. (Responsible: BOC; Timing: MT) *Source: CANADA, INTERNATIONAL MONETARY FUND — EXECUTIVE SUMMARY and chapter content provided in the source PDF.* ### EXECUTIVE SUMMARY __________________________________________________________________________ 7 ### EXECUTIVE SUMMARY ### Macroeconomic landscape and scenario overview - Canadian economy achieved a soft landing after the pandemic but faces rising trade tensions and policy uncertainties. - The Bank of Canada (BOC) began easing monetary policy in June 2024. - Financial Sector Assessment Program (FSAP) adverse scenario: - Time horizon: 2025–27. - Features: deepening geoeconomic fragmentation, protectionism, increased cross-border restrictions, disruption of global production chains, temporary supply shortages, an initial rise in inflation, central banks pausing or reversing interest rate cuts, elevated interest rates in initial years, falling oil prices as growth slows, market valuation corrections, sharp drop in residential real estate prices, equity price corrections, exchange rate depreciation. - Real GDP: cumulative decline of 4.7 percent over the initial two years of the scenario, with a modest recovery in 2027. ### Financial system resilience — overall assessment - The financial system has remained stable amid sizable swings in output, inflation, and interest rates. - Credit performance of deposit-taking institutions (DTIs) is robust; capital and liquidity buffers exceed regulatory minima. - Nonperforming loans have remained at a low level despite a recent uptick. - Nonbank financial institutions (NBFIs) — including insurance, pensions, and investment fund sectors — have weathered shocks well. - Structural household vulnerabilities persist: high debt levels and rising debt-servicing costs compared to G7 peers. ### Household and corporate sector vulnerabilities - Households: - Mortgage Debt Service Ratio (DSR) increases from 8.1 percent in 2024: Q4 to 10.3 percent by end-2026 under the adverse scenario before easing to 8.6 percent. - Mortgage probabilities of default (PDs) under the adverse scenario reach 0.9 percent for uninsured and 1.4 percent for insured loans by end-2027. - Financial strain most evident among renters and lower-income households; delinquency rates in non-mortgage credit increasing. - Households prioritized mortgage repayments through increased savings and disciplined spending; rising unemployment and elevated renewal payments could exacerbate vulnerabilities. - Corporates: - Corporate PDs increase from 0.5 percent in 2024: Q4 to 1.3 percent by end-2027 under the adverse scenario. - Large non-financial corporations generally stable on leverage and liquidity; vulnerabilities concentrated among small and mid-sized businesses, especially in consumer-facing sectors (food services, accommodation, retail). - Expiration of pandemic-era support exposed pockets of weakness; small business insolvencies rose sharply in 2023 before moderating. ### Bank solvency stress test results - Sample and scope: - Seven systemic DTIs analyzed. - Key outcomes: - Average Common Equity Tier 1 (CET1) ratio declines by 2.3 percentage points at the trough under the adverse scenario but remains above regulatory minimums. - At the individual level, the seven systemic DTIs remain well above the 4.5 percent CET1 minimum requirement, and above the 8 percent threshold below which dividend distributions would be affected. - Drivers of CET1 declines: rise in loan losses and increase in risk-weighted assets (RWAs). - Credit risk linkages: - Credit risk and losses incorporate household and corporate sector analyses described above. ### Bank liquidity stress tests - Tests performed: - Liquidity Coverage Ratio (LCR)-based tests over a 30-day horizon. - Cash flow (CF) analysis over a three-month horizon. - Findings: - The seven systemic DTIs are relatively insulated from market-valuation shocks over 30 days but are more vulnerable to significant wholesale deposit outflows. - Cash flow analysis suggests the banking system can withstand large liquidity shocks over a three-month horizon. ### Insurance and pension analysis - Insurance solvency: - Canadian insurers remain solvent under the adverse scenario focused on market and credit risks. - Life insurers: larger variations in capital position but resilient; small cumulative decline in solvency ratios over three years; benefit from higher interest rates and CAD depreciation in the first year, then declining solvency ratios that remain above regulatory thresholds and mostly above or close to internal operational capital targets. - Property and Casualty (P&C) insurers: lower sensitivities to market and credit risks; very resilient in the adverse scenario; relatively larger impact in the first year since benefiting impact from interest rates is lower. - Pensions funding: - Pension plans face deterioration of funding ratios in the adverse scenario but funding levels remain considerably above those seen prior to 2020. - Most scenario shocks are felt in the second year; funding ratios stabilize and increase slightly thereafter for most plans. - Over a quarter of the plans would experience a funding deficit in the third year (which would not trigger any immediate refunding requirement). - Insurance and pensions liquidity: - Bottom-up analysis of a severe adverse scenario with considerably higher short-term interest rates and CAD depreciation: - Margin calls for 19 pension plans and three large life insurers amount to CAD 32 billion, of which 26 billion are due for the pension plans. - CAD 32 billion is around 1 percent of combined balance sheet assets. - Life insurers would rely mostly on sizable holdings of liquid assets. - Pension plans would source a bit more than half of liquidity needs from financing transactions (expiring reverse repos and committed credit lines); about one third from liquid assets. - Many participating entities use contractual arrangements allowing settlement in kind, lowering the need to liquidate assets. ### Interconnectedness and contagion analysis - Scope: - Evaluated funding and credit shock propagation among six D-SIBs and across sectors and borders. - Findings: - Interbank exposures are significant, but spillover risks from bank failures are mitigated by banks’ capital buffers. - Cross-sectoral and cross-border risks are relatively high. - Multiple transmission channels in Canada’s complex and interconnected financial sector imply risks can propagate easily and substantially across sectors. - Implication: vigilant monitoring and analysis are required to identify potential vulnerabilities that could emerge in times of stress. ### Key quantitative highlights (selected) - Scenario horizon: 2025–27. - Real GDP cumulative decline: 4.7 percent over initial two years. - Number of systemic DTIs stress-tested: seven. - Average CET1 decline at trough: 2.3 percentage points. - CET1 minimum requirement: 4.5 percent. - Dividend-affecting CET1 threshold: 8 percent. - Mortgage DSR: 8.1 percent in 2024: Q4 → 10.3 percent by end-2026 → 8.6 percent thereafter. - Mortgage PDs by end-2027 under adverse scenario: 0.9 percent (uninsured), 1.4 percent (insured). - Corporate PDs: 0.5 percent in 2024: Q4 → 1.3 percent by end-2027. - Margin calls in severe liquidity scenario: CAD 32 billion total; 26 billion for pension plans. - Margin calls share of combined balance sheet assets: around 1 percent. ### Policy implications and high-level recommendations - Maintain vigilant monitoring and analysis of cross-sectoral and cross-border transmission channels given high interconnectedness. - Continue monitoring household vulnerabilities (high debt, rising debt-servicing costs), with focus on renters and lower-income households. - Monitor small and mid-sized enterprises in consumer-facing sectors for increased insolvency risk. - Ensure banks retain adequate capital and liquidity buffers given potential rises in loan losses and RWAs under adverse macro scenarios. - Monitor liquidity risks in insurers and pension plans, including preparedness for margin calls and the use of contractual in-kind settlement mechanisms. *Source: CANADA, INTERNATIONAL MONETARY FUND — EXECUTIVE SUMMARY.* ### 1. Establish regular coordination meetings among BOC and regulatory ### 1. Establish regular coordination meetings among BOC and regulatory agencies to discuss stress testing methodologies, results, and implications. ### Financial-sector supervisory and data-strengthening recommendations - Establish regular coordination meetings among BOC and regulatory agencies to discuss stress testing methodologies, results, and implications. - Responsible: BOC, OSFI, AMF - Timing: ST - Continue developing structural models for households and corporates supported by the continuous investment in rich datasets. - Responsible: BOC, OSFI, AMF - Timing: ST - Continue to closely monitor mortgage refinancing risks, household and corporate liquidity buffers, and employment-driven vulnerabilities. - Responsible: BOC, OSFI, AMF - Timing: ST - Start building a credit register for corporate, commercial and consumption loans and continue working towards improving the quality and coverage of Real Estate Secured Lending (RESL) data. - Responsible: OSFI, AMF - Timing: ST - Collect long historical time-series data on default rates across various sectors and asset classes by country of exposure. - Responsible: OSFI, AMF - Timing: ST - Calculate provisions in DTIs top-down stress test exercises in accordance with the International Financial Reporting Standards (IFRS) 9 framework. - Responsible: BOC, OSFI, AMF - Timing: MT - Collect and maintain comprehensive information on IFRS 9 transition matrices and relevant parameters, both prospectively and retrospectively. - Responsible: BOC, OSFI, AMF - Timing: MT - Develop a top-down approach for assessing systemic DTIs’ market risk that evaluates the impact of sudden market shocks, while factoring in the effects of hedging strategies. - Responsible: BOC, OSFI, AMF - Timing: ST - Consider publishing regularly top-down stress test results on an aggregate basis in the Financial Stability Report and/or Staff Analytical Notes. - Responsible: BOC - Timing: MT - Introduce macroprudential stress testing in the insurance sector. - Responsible: OSFI, AMF (in cooperation with BOC) - Timing: MT - Expand supervisory reporting for large pension plans (on investments, derivatives, liquidity, and leverage), and roll out LCR monitoring to all large plans. - Responsible: OSFI, FSRA, FA, PA, BOC - Timing: ST (Note: I—Immediate (within 1 year), ST—Short term (within 1–3 years), MT—Medium term (within 3–5 years)) ### Macroeconomic context and sectoral vulnerabilities (key findings) - The Canadian economy emerged well from the pandemic and achieved a soft landing. - Headline inflation has eased to 2.3 percent, as of March 2025. - Canada trimmed monetary policy rates by 225 basis points to 2.75 percent in March 2025. - By March 2025, Canada's unemployment rate rose to 6.7 percent. - Weaker growth and rising geopolitical uncertainties may weigh on growth, employment, and inflation. - Real GDP growth is expected to soften, mainly due to the US tariffs. - Housing market developments and vulnerabilities: - Home prices surged nearly 60 percent during 2020–22, corrected by about 15 percent in late 2022, and saw a mild rebound of around 5 percent in 2023. - Aggregate house price index used is provided by the Canadian Real Estate Association. - Commercial real estate (CRE) and construction vulnerabilities: - Equifax (December 2024) reported financial trade delinquencies over 60 days in the construction sector rose from 2.9 percent in 2023: Q2 to 3.3 percent in 2024: Q2. - Delinquency rate on asset-based loans more than doubled over the same period. - Large insurance companies and pension funds have an asset exposure of approximately 2–3 percent to the office CRE segment, whereas banks’ exposure stands at around 1 percent. - Bank of Canada analysis (Box 2, Financial Stability Report—2024): approximately 5 percent of large banks’ assets are exposed to the CRE sector, with 0.6 percent in the office subsector. For small and medium-sized banks, CRE exposure is 16 percent, including 1.2 percent in office CRE. Large insurance companies and pension funds have 12 percent and 15 percent of their assets in CRE, respectively, with office CRE accounting for 2.8 percent and 2.5 percent. - Household vulnerabilities: - Debt levels among the highest in the G7; mortgage debt dominates household liabilities. - Household DSR peaked in late 2023 before easing slightly but remains near historical highs. - Mortgage DSR exceeds pre-pandemic levels. - As of early 2025, approximately 60 percent of outstanding mortgages will renew by 2026 and around half of them could be subject to a higher interest rate at renewal. - Variable Rate Mortgage with Fixed Payments (VRMFP) in negative amortization declined from a peak of 31 percent in 2023 to about 12 percent as of September 2024 (representing 2 percent of all mortgages). - In Canada, 80 percent of the residential mortgage portfolio consists of fixed-rate loans, while the remaining 20 percent are variable-rate loans, 75 percent of which are Variable Rate Mortgage with Fixed Payments. - Authorities expect no more VRMFP should reach their trigger rate based on authorities’ estimation after interest rate cuts in early 2025. ### Financial sector structure and banking sector metrics (key statistics) - Canada is the 8th largest financial system in the world by total financial assets. - As of 2024, total assets of financial institutions reached 756 percent of GDP, increasing by 43.3 percent since 2019. - Nonbank financial intermediation accounts for about 60 percent of financial system assets. - Banking sector concentration: - The six largest banks account for 94 percent of the sector's assets; the two largest institutions represent 47 percent. - Including Desjardins, seven Domestic Systemically Important Deposit-taking Institutions (DTIs) hold over 90 percent of all DTIs’ assets. - Balance sheet of those seven systemic DTIs expanded by almost 50 percent during the last five years. - As of 2023, banks’ cross-border exposures in the U.S. represented 29 percent of total assets. - Asset composition (7 systemic DTIs, by end-2024): - Loans constitute 50 percent of total assets. - Residential mortgage loans make up 40 percent of the total loan portfolio. - Loans to Non-Financial Corporations (NFCs) account for 33 percent of the loan portfolio, of which 4.9 percent corresponds to CRE and 7.7 percent are linked to construction and real estate. - Securities represent 24 percent of total assets, primarily government securities (70 percent). - Medium-sized banks (MSB) balance sheets: loans represent 84 percent of total assets. - Funding composition: - Retail and commercial deposits account for 55 percent of funding; MSBs rely more heavily on retail and commercial deposits at 73 percent of liabilities. - Foreign funding accounts for more than 50 percent of total liabilities. - Profitability and asset-quality metrics: - ROA for the 6-DSIBs hovers around 0.7 percent. - Average ROE for Canadian banks is 12 percent as of 2024. - NPL ratio remains low at 0.7 percent. - Mortgage delinquency rates stand at 0.2 percent. - Capital and liquidity buffers: - Average CET1 ratio for the 6 D-SIBs was 13.3 percent as of end-2024. - Average LCR of the 6 largest banks was 125 percent, above the regulatory minimum of 100 percent. *Source: IMF staff compilation from the provided chapter content.* ### 13. The insurance sector is large and interconnected with the rest of the financial system. ### 13. The insurance sector is large and interconnected with the rest of the financial system. ### Market size and structure - Canada’s insurance market ranks ninth globally in terms of written insurance premiums. - Life insurance sector: - Concentration is very high, with the three largest insurers accounting for around 85 percent of the total market. - Mortgage insurance is a small niche but plays an important role in housing finance and is vulnerable to economic downturns and house price declines. - Nonlife insurance sector: - More diversified; motor vehicle and property insurance are the leading subsectors. ### Asset allocation and risk-taking (Life and Non-life insurers) - In 2023, federally regulated life insurers managed total assets of CAD 1,749 billion. - 56 percent are managed as general funds. - 44 percent as segregated funds. - Within general funds (life insurers, 2023): - Bonds: almost 50 percent. - Equity investments: 6 percent. - Real estate: 3 percent. - A substantial portion of “other investments” (e.g., infrastructure) is also present. - Risk behavior and adjustments: - Prolonged low-interest rates in prior years led life insurers to increase risk appetite for higher yielding and less liquid investments. - More recently, some insurers reduced their risk appetite in the CRE sector and accordingly their exposures. - Non-life insurers generally hold more liquid and less risky assets and make very little use of derivatives. - Fixed-income quality: - The share of assets below investment grade is below 2 percent for both sectors combined. - In the non-life sector, less than 1 percent of fixed-income assets are invested below investment grade, compared to 3 percent in the life sector. - The life sector holds a considerably larger share of unrated bonds than the non-life sector—these are mainly private placements. ### Solvency metrics - Solvency ratios (federally chartered insurers, median firm): - Life insurers: reached 172 percent for the median firm in 2023. - Solvency ratios of federally chartered life insurers have been very stable, with the median insurer between 150 and 175 percent since 2019. - Non-life insurers: median insurer reached a solvency ratio of 398 percent in 2023, steadily improving since 2019. - In the federally chartered non-life sector, solvency ratios have been trending upward since 2021, reaching almost 400 percent for the median insurer in 2023. - Notes: Rating breakdown and solvency ratios refer to federally regulated insurers only. ### Pension sector size, allocation, and funding - Assets of trusteed pension funds reached CAD 2.2 trillion in 2023 (75 percent of GDP). - Four-fifths of assets are held by public sector funds. - More than 16,000 plans exist, many extremely small. - The largest public-sector plans (“Maple Eight”) are defined-benefit (DB) schemes. - Investment composition and trends (end-2023 unless stated): - Equities: 37 percent. - Bonds: 26 percent. - Real estate: 11 percent. - Infrastructure: 10 percent. - Over the last two years, the share of real estate and infrastructure increased further, while equities and bonds declined slightly. - Equities, real estate and infrastructure are predominantly invested abroad; more than 80 percent of bond investments are domestic. - Pension plans have increased investment in longer-term (and less liquid) assets and increased leverage using derivatives during prolonged low interest rates. - Funding ratios: - Funding positions of the largest DB plans faced significant pressures during the prolonged low interest rate phase; since 2021, funding positions have considerably improved. - The median funding ratio of a representative sample of pension plans, which had long fluctuated below 100 percent, reached values around 120 percent in 2023/24. - The number of underfunded pension plans has declined substantially. - Median funding ratio based on a sample of around 450 to 500 Canadian pension plans (varying composition). ### Macrofinancial stress scenarios and implications for insurance, pensions, and households - Scenario framework: - FSAP evaluated resilience under two scenarios: a baseline consistent with the October 2024 WEO projections and an adverse scenario consistent with the FSAP Risk Assessment Matrix (RAM). - The adverse scenario is simulated using the IMF’s Global Macro financial Model, disaggregated into forty economies, and spans three years (2025–27). - The adverse scenario features deepening geoeconomic fragmentation, greater protectionism, cross-border restrictions, supply shocks, higher inflation expectations, elevated central bank policy rates in the first years, reduced global growth, higher unemployment, lower commodity demand, and asset valuation corrections. - Key adverse scenario outcomes for Canada: - Real GDP: cumulative decline of 4.8 percent over the initial two years of the scenario, comprising contractions of 2.4 percent in both 2025 and 2026. - The two-year cumulative loss represents a 2.7 standard deviation shock from the mean of the historical distribution. - Unemployment: rises, reaching a level of 9.2 percent at the peak. - Residential real estate prices: drop by 25 percent, driven by high interest rates, increased unemployment, reduced household income, and a sharp decline in immigration. - Inflation: increases moderately, by about 1pp. - Short-term rates: increase 60 bps (driven by high inflation expectations). - Long-term rates: increase. - Exchange rate: CAD depreciates by about 10 percent. - Stock prices: fall by 33 percent. - Nominal wage growth: declines by 0.7 percentage points. - Oil prices: slump by 33 percent. - Household vulnerabilities: - Canada’s household sector is the most indebted among G7 economies, with household debt levels exceeding 100 percent of GDP and 170 percent of disposable income as of 2024. - Residential mortgages constitute 74 percent of total credit liabilities of households (2024: Q4) and account for a substantial portion of DTIs’ lending portfolios. - Mortgage arrears are below pre-pandemic levels, reflecting household adaptation to higher interest rates through precautionary savings and expenditure adjustments, but material household risks remain if a deep macroeconomic downturn occurs with sharply rising unemployment and stagnating income growth. *Source: IMF staff calculations and analysis as presented in the chapter titled "The insurance sector is large and interconnected with the rest of the financial system."* ### 23. There is heterogeneity across different income groups and debt instruments. ### 23. There is heterogeneity across different income groups and debt instruments. ### Heterogeneity in repayment behavior and emerging vulnerabilities - Canadian households with mortgages generally prioritize mortgage repayments over non-mortgage credit obligations, reflecting strong incentives to maintain homeownership under a full-recourse system in most provinces. - Inflationary pressures, a slower-than-expected pace of interest rate cuts, and increases in debt service payments as mortgage rates reset from pandemic lows could exacerbate financial vulnerabilities, particularly for borrowers with limited liquidity buffers. - Delinquency rates have increased more pronouncedly in non-mortgage credit instruments—especially credit cards and auto loans—particularly among renters and lower-income borrowers. - Since 2017, uninsured mortgages have dominated Canada’s residential mortgage market (around 75 percent as of 2024Q4), growing more rapidly than insured loans, particularly during the pandemic, and continuing to outpace them despite the recent market slowdown. - The rise in uninsured mortgages reflects regulatory constraints that render homes valued over CAD 1 million ineligible for mortgage insurance and require a minimum 20 percent down payment; uninsured lending is more prevalent among higher-value properties, particularly single-detached homes. ### Wealth distribution and income effects (2024: Q2) - Top 20 percent of households held about 68 percent of total net worth, averaging 3.4 million CAD per household. - Bottom 40 percent accounted for only 2.8 percent of total net worth, averaging about 70 thousand CAD per household. - Post-pandemic wealth disparities widened: financial asset gains benefited higher-income households; lower- and middle-income households faced rising debt burdens. - Higher interest rates in 2023: - boosted investment earnings for wealthier households; - strained disposable income for middle-income households as interest payments outpaced wage growth. - Lowest-income households saw real income gains in early 2024, but these gains were offset by rising mortgage debt. - Debt-to-income (DTI) ratios are markedly higher for the lowest-income quintile, which has experienced the largest increase in indebtedness since 2020. ### Debt service burdens and recent trends - Overall household debt service ratio (DSR): - peaked at 15.1 percent in 2023: Q4; - moderated to 14.7 percent in 2024: Q3. - Mortgage DSR for all households: - rose above the non-mortgage DSR post-2022 for the first time since mid-1990s; - reached a record 8.2 percent in 2023: Q2 before easing slightly to 8.0 percent in 2024: Q3. - Negative amortization loans accounted for around 2 percent of all mortgages as of 2024: Q3. - Employment dynamics are a crucial determinant of mortgage repayment capacity and arrear rates. ### Mortgage risk analysis: methodology - The FSAP conducted a quantitative mortgage risk analysis integrating: - a structural simulation method (bottom-up simulation of DSR) using loan-level OSFI RESL data; - a Bayesian Model Averaging (BMA) econometric framework to project PDs using aggregate household statistics from Statistics Canada and historical “Actual PDs” from banks’ public pillar 3 disclosures covering 2014 through 2024. - Structural simulation details: - initialized household financial variables to reflect end-2024 economic conditions; - used a bootstrap approach to model employment dynamics aligned with projected unemployment dynamics and macro trends; - adjusted household income based on simulated employment outcomes and wage growth expectations to produce DSR projections. - BMA framework: - systematically selected relevant predictors for mortgage PDs; - key variables included GDP growth, inflation, interest rates, labor market dynamics, and external macroeconomic conditions; - mortgage DSRs from the first stage were anchored to end-2024 aggregate Statistics Canada statistics. - Data limitation: projections do not incorporate new mortgage originations after late 2023 (based on the RESL dataset received); the decline in DSR primarily reflects amortization and improving repayment capacity among existing borrowers. ### Projection results: baseline vs adverse scenarios - Mortgage DSR projections: - Baseline: mortgage DSR declines from 8.1 percent in 2024: Q4 to 6.7 percent in 2027Q4, reflecting steady income growth and stable debt burdens. - Adverse: DSR peaks at 10.3 percent in 2026: Q4 before easing to 8.6 percent by 2027: Q4, driven by higher interest rates, weaker income growth, and rising unemployment. - Income-group impacts under adverse scenario: - Low-income households face the sharpest DSR increase, exceeding the baseline by over 2 percentage points in 2025–26. - Middle-income borrowers see a moderate rise with less volatility. - High-income groups remain largely insulated. - Mortgage Probability of Default (PD) projections under the adverse scenario: - Uninsured loans PD reaches 0.9 percent. - Insured loans PD reaches 1.4 percent. - Primary drivers of mortgage defaults: - Unemployment is the key risk driver for uninsured loans. - Wage growth is the main factor for insured loans (reflecting higher LTV ratios and lower income profiles typical of insured borrowers). - Note on interpretation: mortgage PD projections are based on a short time series for the panel BMA estimation due to unavailability of longer historical data. ### Policy implications and monitoring priorities - Household sector remains overall resilient under the adverse scenario, but vulnerabilities persist among lower-income borrowers and in non-mortgage credit instruments. - Critical areas for policy focus: - Close monitoring of refinancing and default risks across various household debt instruments. - Emphasis on savings accumulation and liquidity buffers, particularly for lower-income borrowers with limited financial flexibility and vulnerability to unemployment shocks. - Recommended further analytical and supervisory actions: - Authorities should further develop structural models (such as the latest BOC’s Household Risk Assessment Model (HRAM)) leveraging rich data access to micro-level household information and incorporating counterfactual analysis to assess vulnerabilities more comprehensively, particularly in the context of low arrears in past cycles. *Source: IMF Staff analysis and FSAP findings as presented in the supplied content.* ### 39. Indicators of firms’ profitability and liquidity further illustrate the uneven financial ### Indicators of firms’ profitability and liquidity further illustrate the uneven financial recovery ### Firm profitability and liquidity trends - Return on Assets (ROA) has remained stable for the median firm and top-performing firms; profitability has deteriorated among firms in the lower quartile, particularly in the post-pandemic period. - Liquidity ratio (current assets over current liabilities): - Has remained stable for large firms, reflecting stronger balance sheets and financing access. - Has been more volatile for medium-sized firms, indicating greater exposure to financial constraints and less sufficient buffers. - Aggregate implication: improving aggregate financial risk profile but continued vulnerabilities among lower-profitability and highly leveraged firms in a tightening financial environment. ### Debt-at-risk dynamics and sectoral divergence - Post-pandemic trends: - Debt-at-risk has steadily declined over time in trade, real estate, professional activities, information and communication, and accommodation and food services, reaching lower levels compared to historical trends. - Mining, manufacturing, and agriculture have exhibited higher and more volatile debt-at-risk levels, reflecting sector-specific vulnerabilities (exposure to commodity prices, global supply chain disruptions, capital-intensive financing). - Note on alternative measure: Using the BOC measure (ICR based on EBITDA rather than EBIT and alternative pre-filtering) results in a level shift of ICR to a lower range. ### Firm-level regressions: liquidity buffers, debt structures, and default risk - Firm-level fixed-effects regressions consistently identify liquidity constraints as a key driver of financial distress. - Firms maintaining higher cash buffers experience significantly lower probabilities of default. - A positive interaction term indicates that lower profitability increases the importance of cash buffers in mitigating default risk. - Industry heterogeneity: - Interaction between ICR and cash buffers is particularly pronounced in trade and services (thinner margins, more sensitivity to liquidity shocks). - Capital-intensive sectors are more exposed to leverage/refinancing risks; lower-margin industries depend more on liquidity buffers. - Finding consistency: Similar dynamics observed using the Bank of Canada (BOC) filter and measures; key findings remain broadly consistent. ### BMA projections of corporate Probability of Default (PD) and sector vulnerabilities - Baseline scenario: corporate PDs remain stable at approximately 0.5 percent in early 2025 before gradually declining. - Adverse scenario: corporate PDs rise to 1.3 percent by late 2027. - Sectors with pronounced projected PD increases: mining, quarrying, oil and gas, construction, and real estate (high sensitivity to commodity prices and capital-intensity; elevated refinancing risks). - Macro-financial factor heterogeneity: - Corporate spread is the most important macro factor driving overall PD; matters most for wholesale trade, natural resources, and construction and real estate. - FX volatility is most relevant for mining, quarrying, oil, and manufacturing. - Unemployment shocks have the strongest effect on natural resources, retail trade, and mining. ### Policy priorities and monitoring recommendations - Focus areas: 1. Continue close surveillance of corporate liquidity indicators and debt-servicing burdens, particularly in capital-intensive and commodity price-sensitive sectors. 2. Keep assessing the financial position of SMEs to gauge potential spillover effects on employment and financial stability. 3. Evaluate the potential impact of trade disruptions on corporate sector financial health and implications for labor markets. 4. Continue integrating firm-level corporate default modeling into financial stability assessments to better capture sectoral risks and firm-specific vulnerabilities, particularly under trade uncertainty and global economic volatility. ### Bank solvency stress test: scope and framework - Coverage and data: - Exercise comprised the seven systemic DTIs (the six D-SIBs and Québec’s domestic systemically important institution (D-SIFI)). - Those seven systemic DTIs hold over 90 percent of the overall DTI sector’s assets. - Cutoff date for the exercise: end of fiscal year 2024. - Main data source: supervisory returns collected by OSFI and the Autorité des Marchés Financiers (AMF). - Approach: - Standard FSAP top-down solvency stress-test with a three-year horizon under a baseline and an adverse scenario. - Channels considered: credit risk, interest rate risk, and market risk. - Link from macro scenario to financial statements via satellite models for credit risk and P&L components. - Balance sheet assumed static (asset allocation and funding composition unchanged from cutoff date). - Dividends: under positive profits and if Capital Conservation Buffer (CCoB) is not breached, payout ratio set at observed cutoff ratio; if CCoB breached (CET1 below 8 percent), restrictions aligned with regulatory framework; if net profits negative, no dividend payout assumed. - Capital buildup assumed only via retained earnings (no new equity issuance). - Post-stress capital formula: CCr* = Cc0 + Issued* - RWA* (as expressed in the source text). ### Capital requirements and hurdle rates used - Baseline hurdle rates: - Pillar 1 CET1: 4.5 percent - Capital Conservation Buffer (CCoB): 2.5 percent - D-SIBs surcharge: 1 percent - Domestic Stability Buffer (DSB): 3.5 percent (not required for Quebec’s D-SIFI) - Adverse scenario: DTIs allowed to deplete the CCoB and the DSB; other requirements remain in place. - Note: Pillar 2 capital requirements are treated as a supervisory expectation (not legally binding in Canada). ### Credit risk modeling and provisioning - Provisions calculated as expected credit losses (ECL) for all asset classes/sectors with key parameters PD, LGD, performing and non-performing exposures. - For corporate loans: - PD projections generated using BMA-estimated models linking reconstructed historical PDs (provided by the BOC) to macro variables. - Models estimated for 11 corporate sectors; US macro variables included among explanatory variables due to limited historical proxy PDs by sector/geography. - For mortgage loans: - PD projections generated using a Panel BMA approach; structural model projects DSR using loan-level data and Panel BMA projects PDs conditional on macro variables (including aggregate DSR). - Other consumer loans’ PDs (HELOCs, credit cards) linked to the mortgage PD path via satellite models. - Mapping to banks: - Aggregate PD paths mapped at bank level using standard scores (z-scores) to ensure projected PDs remain within the [0,1] range. - LGD treatment: - LGD paths for mortgage and CRE loans linked to house price paths via the model provided in the source text. - Downturn LGDs updated only if stressed LGDs exceed starting levels; otherwise kept constant. - Credit risk RWAs: - For standardized approach (SA), densities at the cut-off point assumed constant over the scenario horizon. - For Internal Ratings Based exposures, Basel formulas used; through-the-cycle PDs calculated using a moving-average–like formula based on projected PiT PDs. ### Interest rate risk and repricing structure - Net Interest Income (NII) projection: - Structural model simulates repricing/origination at prevailing rates for new business and time until next repricing/maturity. - Assumes distribution of assets/liabilities across repricing buckets remains constant; interest on nonperforming exposures excluded; no relative composition changes assumed. - DTI-specific inputs: initial exposures generating interest income/expenses; repricing/maturity profile (from OSFI I3 and AMF data); initial effective interest rates; projection of interest rates on new business. - Repricing structure (assets and liabilities): - Non-mortgage loans: significant majority reprice within one year. - Mortgage loans: - Approximately 50 percent reprice within one year. - 34 percent reprice within one to three years. - Remaining mortgage loans have repricing terms beyond three years. - Over 50 percent of the mortgage portfolio comprises variable rate loans and fixed-rate loans with terms shorter than 5 years. - Deposits (liabilities): - About 61 percent of personal deposits reprice within one year. - About 77 percent of non-personal deposits reprice within one year. *Source: IMF staff analysis in the cited chapter.* ### 58. Interest rates on new business were projected using simple passthrough regressions. ### Interest rates on new business were projected using simple passthrough regressions. ### Interest-rate projection methodology - Interest rates on new mortgage and non-mortgage lending were projected using simple passthrough regressions based on time series data for the aggregated banking system from OSFI’s “Report on New and Existing Lending (A4)”. - Deposit rates modelled using time series data on Guaranteed Investment Certificates (GICs) rates provided by the BOC. - For securities and other interest-earning assets, a pass—through rate of 100 percent was assumed. - Estimated lending—and deposit betas (as presented in the source) for Delta Interest Rates and Delta GIC rates: - Non-mortgage loans Delta Interest Rate: 0.955*** (0.0846) - Mortgage loans Delta Interest Rate: 0.824*** (0.0809) - Delta GIC rates: 0.907*** (0.156) - Constants: -0.0302, -0.0743, -0.0424 (standard errors shown as (0.0796), (0.0761), (0.179)) - Observations: 555, 537 - R-squared: 0.7060, 0.6620, 0.492 ### Modelling of other profit and loss components - Satellite models estimated for non-interest income ratio (NIIR) and non-interest expense ratio (NIER). - Key drivers: - Non-interest income: equity price and interest rates (market activity influences fees and commissions). - Non-interest expense: inflation (affects operational costs including salaries, rent, utilities). - Scenario behavior: - Baseline: NIIR maintains its recent upward trajectory; NIER decreases consistent with baseline inflation path. - Adverse: NIIR experiences a temporary decline in line with decreasing equity prices before resuming an upward trend; NIER increases consistent with adverse inflation path. - Aggregated P&L ratio paths are mapped at the DTI level based on starting points of the ratios. ### Market risk modelling approach - Securities portfolios account for 24 percent of total assets. - Accounting composition of securities portfolio: - FVPL: 53 percent - AC: 26.5 percent - FVOCI: 20.3 percent - Approximately 22.8 percent of total securities are issued or guaranteed by the Government of Canada, Canadian provinces, or municipal and school corporations. - Market risk module for sensitivity analysis: - Captures valuation changes of debt securities due to changes in risk-free interest rates and credit spreads. - FVPL and FVOCI: mark-to-market estimation. - Modified duration approach used to reevaluate exposures as function of residual duration, relevant bond yield and stressed spreads. - Stressed spreads consistent with macroeconomic scenario and equal to those applied for the insurance stress test. - For conservatism, in absence of information, existing hedges assumed ineffective during the scenario horizon. - Accounting treatment implications: - FVPL losses/gains considered realized, affect net profits and are subject to taxation and dividend payout. - FVOCI unrealized losses/gains affect capital through other comprehensive income. ### Solvency stress test results - Aggregate CET1 capital ratio trajectory: - Starting CET1 in 2024: 13.7 percent - Baseline projection by 2027: 15.7 percent - Under the adverse scenario, CET1 for the seven systemic DTIs declines by 2.3 percentage points at the trough but remains above regulatory minimums. - Individual-level outcomes: - All systemic DTIs remain above the 4.5 percent CET1 minimum requirement and above the 8 percent threshold below which dividend distributions would be affected. - Main drivers of CET1 changes: rise in loan losses and increase in RWAs; to a lesser extent decreases in NII and non-interest income and increases in non-interest expense. - Sensitivity analyses conducted: 1. Market risk: valuation risks of securities portfolio due to changes in risk-free rates and credit spreads. 2. Sectoral credit risk: higher increases in PDs for manufacturing and commodity sectors. 3. Conservative PD path: more conservative PDs across all corporate and household loan portfolios. - Sensitivity outcomes: - Market risk sensitivity: average CET1 for the 7 systemic DTIs falls to 10.4 percent in 2026; all institutions remain above the 8 percent threshold. Losses are material only in the first year; gains contribute positively in outer years as rates normalize. Hedging assumed ineffective (upper-bound of potential losses). - Higher PDs for manufacturing and commodity sectors: additional CET1 drop of 40 basis points relative to adverse scenario. - Conservative PDs across all portfolios: additional CET1 drop of 1.7 percentage points relative to adverse scenario, reaching 9.8 percent in 2027. - Under all sensitivity analyses, institutions remain above the 8 percent threshold. ### Liquidity stress testing overview and LCR-based test - Liquidity analysis assessed resilience of the 7 systemic DTIs to funding shocks and market-driven stress via: - LCR-based test (30-day horizon). - Cash Flow (CF) / Net Cumulative Cash Flow (NCCF) analysis (one week to one year horizons, emphasis on three months). - LCR context and calibration: - By October 2024, aggregate LCR for the 7 systemic DTIs was 136 percent; no entity had ratio below the regulatory minimum. - LCR adverse scenarios combine higher haircuts on HQLA and higher run-off rates on outflows. - Test combined three HQLA haircut scenarios with four outflow scenarios for a total of 12 combined scenarios: - Regulatory haircut (OSFI and AMF aligned to Basel standard) and two progressively severe haircut scenarios. - Regulatory outflow scenario (OSFI and AMF) plus: - Retail outflow scenario: increased run-off rates for retail funding; wholesale run-offs as regulatory. - Wholesale outflow scenario: increased run-off rates for wholesale funding; retail run-offs as regulatory. - Combined scenario: stressed run-off rates for both retail and wholesale. - Scenarios calibrated using relevant historical episodes including the banking turmoil in March-2023 and concurrent FSAPs. - Run-off principles applied: - More informed/sophisticated depositors withdraw funding more rapidly than less informed ones (wholesale run-offs higher than retail). - Insured funding sources have lower run-off rates than uninsured funding sources. - Data inputs: - DTI-level tests use OSFI and AMF regulatory data from LCR and NCCF returns as of October 2024. ### Recommendations (summarized) - Enhance interagency collaboration in stress testing among BOC, OSFI and AMF: establish regular coordination meetings, continue standardization of regulatory returns (liquidity templates). - Develop structural models for credit risk to better capture nonlinear mortgage-default dynamics and complement econometric corporate credit loss estimation with corporate stress tests; invest in comprehensive household and corporate datasets and collect long historical series on default rates or proxy PDs. - Develop an independent top-down market risk methodology to assess sudden market shocks and account for hedging strategies; collect information on market sensitivities (e.g., “the greeks”). - Continue building credit registers for corporate, commercial and consumption loans and improve RESL data quality and coverage (e.g., mortgage loans granted outside Canada) to enable granular analyses and counterfactuals tied to prudential policies. - Compute provisions in top-down stress testing aligned with the IFRS9 framework; collect and maintain comprehensive IFRS 9 transition matrices and relevant parameters. - Consider regularly publishing top-down stress testing results in the Financial Stability Report and/or Staff Analytical Notes to enhance transparency. *Source: IMF staff calculations.* ### 75. The results of the LCR-based stress tests indicate that the Canadian systemic DTIs are ### 1canea2025008-source-pdf - 75. The results of the LCR-based stress tests indicate that the Canadian systemic DTIs are ### LCR-based stress tests: aggregate and individual results - Aggregate findings: - The Canadian systemic DTIs are liquid and broadly resilient to sizable funding outflows over a 30-day horizon. - Under all of the individual retail scenarios, the aggregate LCR remains above the regulatory minimum of 100 percent. - In the wholesale and combined scenarios, the aggregate LCR falls somewhat below the 100 percent threshold. - Individual institution findings: - Three systemic DTIs remain resilient under all wholesale scenarios. - Four systemic DTIs fall below the 100 percent threshold in some wholesale scenarios, though not by large amounts. - Most common reasons for breaching the threshold: lower starting LCRs and greater exposure to wholesale outflows. - Figure and data notes: - Figure 23 and Figure 24 present LCR by currency and the LCR-based stress test design across 12 scenarios combining three market scenarios with four run-off rate scenarios. ### LCR analysis by currency - Currencies analyzed: CAD, US dollars (USD), the British Pound (GBP), and the EUR. - As of October 2024: - Aggregate LCR was 131 percent in CAD. - Aggregate LCR was 120 percent in USD. - Currency-specific sensitivities: - The CAD LCR has greater sensitivity to retail deposit outflows; the aggregate CAD LCR drops somehow below 100 percent for the two retail scenarios with higher haircuts. - The USD LCR shows higher sensitivity to wholesale outflows. - Regulatory note: - OSFI monitors the LCR by currency but does not prescribe a regulatory minimum limit. ### LCR-based stress test design (selected parameters) - Stress test structure: - Three market scenarios combined with four run-off rate scenarios for a total of 12 scenarios. - Haircuts / Run-off rate examples (as presented in the source): - Level 1 assets: Regulatory 100%; Haircut 1 95%-100%; Haircut 2 90%-100%. - Level 2A assets: Regulatory 85%; Haircut 1 75%-80%; Haircut 2 65%-75%. - Level 2B assets: Regulatory 50%-75%; Haircut 1 40%-70%; Haircut 2 30%-70%. - Retail and wholesale run-off rate illustrative ranges (from table): - Stable Deposits: 3-5%; Less-Stable Deposits: 10%–20% depending on scenario. - Wholesale Liabilities: examples include 3-5%, 15%, 30%, 40-60% across scenarios. - Note: The stress test combines variations in haircuts and run-off rates to generate the 12 scenario outcomes. ### Cash Flow (CF) analysis: methodology and indicators - Focus and scope: - The CF assesses the seven systemic DTIs’ liquidity risk using two indicators: the cumulative net funding gap and the counterbalancing capacity. - Definitions: - Net funding gap: sum of differences between cash inflows and outflows across all time buckets within a given horizon; positive inflows from the loan portfolio are excluded. - Counterbalancing capacity: cumulative value of liquid assets that banks can liquidate under stress at reasonable prices (mainly cash resources and securities). - Liquidity shortfall arises when counterbalancing capacity is insufficient to meet the net funding gap. - Assumptions: - Contingent credit lines are assumed to be partially triggered, increasing outflows. - The sum of contractual cash inflows/outflows from USD FX and cross currency swap assets/liabilities over all time periods is recognized as an inflow/outflow in week 1. ### Cash Flow scenarios and calibration - Scenario design: - Analysis considers a range of 20 scenarios of increasing severity, affecting outflows and counterbalancing capacity. - Scenarios are based on a linear grid of scenario severities across factors (run-off rates, haircuts) ranging from a mild to an aggressive scenario. - For each scenario, a complete collateral revaluation is conducted, encompassing collateral utilized for both repo and reverse repo transactions. - Role of CF: - CF provides a granular, balance sheet-based approach complementing the LCR and offers insights into potential vulnerabilities to support policy decision-making. - Reference: - Table 5 and Appendix VI contain full scenario details and granular parameters. ### Cash Flow results: horizons, scenarios, and magnitudes - General horizon results: - The systemic DTIs can withstand liquidity shocks over a three-month horizon under most CF scenarios. - Under the milder scenario, the 7 systemic DTIs maintain liquidity surpluses for horizons shorter than six months. - Under the most aggressive scenario, the 7 systemic DTIs maintain liquidity surpluses for horizons equal or shorter than one month. - Three-month horizon, 20 scenarios: - On average, the 7 systemic DTIs would maintain liquidity surpluses even under more severe scenarios. - Shortfalls occur only in the three most aggressive scenarios (scenarios 18–20). - Under the most aggressive scenario, aggregate shortfalls amount to 237 billion CAD, which represent 12 percent of the initial counterbalancing capacity. - Given the severity of the most aggressive scenario, this impact is characterized as moderate. - Institution-level heterogeneity: - One institution starts to experience shortfalls in scenario 12. - Two institutions maintain surpluses under all scenarios considered. - Visual: Figure 25 shows liquidity surplus/shortfall by D-SIFI across scenarios (values in billion CAD, across scenarios 1–20 for Bank 1–Bank 7). ### Interconnectedness and contagion analysis: objectives and approach - Objective: - Illuminate potential vulnerabilities and weaker links within the Canadian financial system by identifying interconnectedness and spillover pathways and assessing systemic risk and contagion potential. - Data sources: - Interbank cross-holding data from the Bank of Canada (BOC) covering the six D-SIBs and aggregated data for all other banks. - Cross-sectoral and cross-border exposures assessed using “from-whom-to-whom” statistics from Statistics Canada. - Model and simulations: - Espinosa-Vega Sole (EVS) model applied to simulate network contagion via credit shocks and funding shocks through bilateral linkages, with iterative tracking of contagion failures (domino effects). - Funding shock simulation includes cases where liquidity surplus is sufficient to meet funding shortfall and where liquidity surplus is insufficient and the remaining shortage must be matched by fire-sale proceeds. - Stylized balance sheet identity used for bank i (as presented in source): - Sum_j x_j i j + a_i = I_i + b_i + W_i + Sum_j x_i j j - Notation: x_ji = bank i loans to bank j; a_i = other assets; c_i = capital; b_i = long-term and short-term borrowing; d_i = deposits; Sum x_ij = bank i borrowing from bank j. ### Contagion and vulnerability indices, model parameters - Indices produced: - Index of Contagion: averages losses of other Banks/Sectors due to the failure of a Bank. - Index of Vulnerability: averages loss of a Bank/Sector due to the failure of other Banks/Sectors. - Parameter assumptions (Table 6): - λ = 0.4 and 1.0 (40 and 100 percent LGD on exposures). - ρ = 0.35 (Share of lost funding that is non-replaceable). - δ = 0.5 and 1.0 (50 and 100 percent discount on asset sales). - Note: 40 percent loss-given-default rate and 50 percent discount rate are applied for the cross-sectoral and cross-border estimates; results sensitive to these assumptions but relative weight of systemic importance remains robust. ### Interconnectedness key findings (overview) - The Canadian financial system is characterized by high interconnectedness: - Between banks and households. - Between households and NBFIs. - Between NBFIs and external sectors. - These interconnections significantly influence systemic risk and underscore the importance of monitoring interbank relationships and cross-sectoral exposures. - Visual: Figure 27 maps financial system structure and interconnectedness; NBFI in the right chart includes pension funds, insurance corporations, investment funds and other depository corporations (ODC); Govnt includes general government, social security funds and government business services. *Sources: OSFI, AMF and IMF staff calculations.* ### 84. The analysis reveals that the average losses due to credit and funding shocks could ### 1canea2025008-source-pdf - 84. The analysis reveals that the average losses due to credit and funding shocks could ### Interbank contagion and vulnerability - The analysis reveals that the average losses due to credit and funding shocks could reach approximately 5 percent of capital across various banks (Figure 28). - Significant interbank exposures exist, but spillover risks associated with bank defaults are cushioned by banks’ robust capital levels. - Contagion and vulnerability indices vary across banks, indicating differing risk profiles among institutions. - Financial System Structure, end-2024 (total assets in percent of GDP): - Banks, 2 33.7 - Othe r Depository Corporations, 27.5 - Insurance corporations, 43.8 - Pension funds, 88.6 - Social Se cur ity, 34.6 - OFIs, 2 04.1 - Investment Funds, 1 23.5 ### Cross-sectoral contagion and vulnerability - NFCs and the external sector exhibit the highest contagion levels but maintain relatively low vulnerability. - Households, banks, and NBFIs are less contagious but more vulnerable, driven by high household debt and sensitivity to market fluctuations. - Policy-relevant implications: - For NFCs and the external sector: maintain strong defenses through prudent financial practices and diversification. - For households, banks, and NBFIs: focus on reducing debt levels, enhancing liquidity management, and improving resilience to market volatility. - Note on interpretation: The estimated loss of capital should be seen as indicative rather than definitive due to inherent uncertainties in modeling financial systems. The definition of "default" assumes total equity exhaustion, a scenario unlikely in Canada given robust capital buffers. ### Risk transmission channels - Cross-border risks are significant, with vulnerabilities to external shocks affecting various sectors (Figure 30). - Most domestic Canadian sectors appear very vulnerable to external shocks (except for investment funds, insurance companies, OFI and NFC). - The early impact of external shocks affects banks, pension funds, social security funds and households. - Household vulnerability can transmit to other sectors primarily via: - Depository corporations - Pension funds - Social security funds - Monitoring household debt levels and their impact on overall financial stability is crucial. ### Insurance and pension risk analysis — scope and methodology - FSAP performed solvency stress tests and liquidity risk analyses for insurers and pension funds. - Insurance sector: - Top-down solvency stress test covered insurers under federal supervision and Québec-chartered insurers. - Sample: eight life insurers and 17 P&C insurers, covering around 90 percent and 50 percent of each sector, respectively, in terms of assets. - Life sector highly concentrated: three large entities and five considerably smaller insurers. - Pension sector: - Solvency analysis for 40 DB and hybrid schemes supervised by OSFI and FSRA. - Bottom-up liquidity risk analysis covered a sample of three life insurers and 19 pension plans, including six of the “Maple Eight” public pension plans—covering more than 50 percent of pension assets. - Scenario design: - Baseline and adverse scenarios follow macrofinancial scenarios used for the banking sector stress test, with adjustments for insurers’ balance sheets. - Adverse scenario includes shocks to risk-free interest rates, equity and property prices, and credit spreads of corporate and sovereign bonds (Table 8). - Additional sensitivity tests (single-factor, instantaneous) were run and not added to adverse scenario results: - Equity: A decline in stock markets by 40 percent (20 percent for preferred shares). - Currencies: An increase in the external value of the CAD by 10 percent. - Modeling assumptions: - Top-down solvency stress test projects valuation changes and solvency coverage ratio over a 3-year horizon. - Shocks applied to investment assets and insurance liabilities as of June 2024 and for June 2025 and June 2026. - Haircuts applied to market value of investments; liabilities (except segregated funds) approximated through a duration-based approach. - Assumed insurers maintain asset allocation over projection horizon with annual re-balancing; interests and dividends unchanged. - Insurance premiums, claims and expenses assumed to change with consumer prices in the scenario. - Capital requirement components for market risk and credit risk recalculated after stress; other components assumed unchanged except diversification benefit adjustment. ### Stress test specification and key scenario shocks (as used) - Interest rates (risk-free rates, bps) and Credit spreads (bps) shocks applied over Y1, Y2, Y3. - Stocks (%) shocks and Exchange rates (%) shocks specified. - Commercial real estate (%) and Other (%) shocks specified. - Examples of numeric scenario elements (as presented): - Canada Stocks: 6,5% (baseline) and -17,3% (adverse, Y1), -18,1% (adverse, Y2), 14,3% (Y3) - United States Stocks: 7,4% (baseline) and -24,5%, -29,5%, 22,8% (Y1–Y3 adverse/baseline) - CAN/USD exchange rate changes: -0,6%, -0,5%, -0,7% (baseline) and 4,7%, 4,0%, -3,1% (adverse) - (Full tabulated shocks provided in the source for interest rates, spreads, stocks, FX, CRE, and other asset classes.) ### Results — valuation changes (life and P&C insurers) - Life insurers: - Baseline: investment assets increase by 2 to 3 percent per year in first two years and slightly less in year 3; liabilities increase at a lower rate than investments. - Adverse: strong decline in both investments and liabilities in year 1 and reversal in year 3. - Year 2: decline in asset prices (5 percent for the median life insurer) and an increase in liability values (3 percent). - P&C insurers: - Lower sensitivities to market and credit risks; investment values and liabilities fluctuate less than life insurers. - Adverse scenario: cumulative investment decline of 12 percent in the first two years for median P&C insurer; liabilities remain largely stable (-1 percent in year 1 and +1 percent in year 2 for the median firm). - Year 3: investment values rise by 10 percent; liabilities increase by 2 percent. ### Results — available capital changes - Life sector: - Median life insurer: available capital increases by 15 percent in year 1, then declines by 18 percent in year 2 and by 9 percent in year 3 under the adverse scenario. - P&C sector: - Available capital declines by 15 percent in year 1 (text also reports declines of 8 and 4 percent in years 1 and 2 respectively in the context of a median description) and then increases by 19 percent in year 3 for the median P&C insurer. - Drivers: - Interest rate shocks have the most pronounced impact on available capital, especially for life insurers. - Equity shocks and, to a lesser extent, credit spreads and real estate are material drivers in the life sector. - FX shock has a very small benefitting impact in years 1 and 2 for P&C sector. ### Results — solvency ratios (median firms) - Baseline: - Life insurers’ solvency ratios increase from 136 to 161 percent cumulatively over three years. - P&C insurers’ solvency ratios increase from 259 to 279 percent cumulatively over three years. - Adverse scenario: - Life insurers: solvency ratio for the median firm initially increases from 136 to 168 percent in year 1, then declines to 142 percent in year 2 and to 121 percent in year 3 (cumulative decline of 15 percentage points). - None of the life insurers sees solvency ratio fall below the regulatory threshold; some insurers remain above or close to internal operational capital targets. - P&C insurers: median solvency ratio falls from 259 to 222 percent in year 1, to 218 percent in year 2, and recovers to 237 percent in year 3. *Source: IMF staff calculations.* ### 101. The sensitivity analysis shows potential vulnerabilities related to equity exposures, ### 101. The sensitivity analysis shows potential vulnerabilities related to equity exposures, ### Insurance sensitivity analysis: equity and FX exposures - Canadian insurers hold material investments in stocks; an instantaneous 40 percent decline in stock markets would cause solvency ratios to decline by 19 and 23 percent in the life and P&C sector, respectively. - Net currency exposures are marginal due to close matching and hedging of currency risks on asset and liability sides. - A 10 percent appreciation of the CAD: - For the median life insurer, the solvency coverage ratio would decline by less than 1 percentage point. - For the median P&C insurer, the solvency coverage ratio would decline by 2 percentage points. - Summary caption (Figure 36): Life insurers would see their solvency ratio declining significantly if equity prices dropped, while the sensitivity to the exchange rate is marginal. P&C insurers have a slightly higher sensitivity to exchange rates, but for the median firm the solvency rate would only decline by 2 percentage rates if the CAD appreciates by 10 percent. - Source of calculations: IMF staff calculations based on OSFI and AMF data. ### Box 1 — Mortgage Insurance Risk Analysis: scenario and impacts - Scope: Top-down risk analysis for the three Canadian mortgage default insurers combining the adverse macrofinancial scenario with higher defaults on insured mortgage loans. - Adverse scenario assumptions: - House prices decline by almost 25 percent in the first year of the projection horizon. - Probabilities of default more than double by year 2 compared to the baseline projection. - Starting position: - Pre-stress solvency ratio of mortgage insurers: 183 percent (weighted average). - Around 63 percent of insured volume have an LTV ratio below 70 percent. - 38 percent of insured volume have an LTV ratio below 50 percent. - Delinquency rates between 0.1 and 0.2 percent over the last two years with a slight uptick towards end-2024. - Combined ratio of the sector was below 15 percent (weighted average). - Returns on equity ranged from 18 to 46 percent in 2023. - Adverse scenario quantified incurred claims: - CAD 290m in 2025 - CAD 520m in 2026 - CAD 450m in 2026 (source text lists 2026 twice) - Combined ratios in the adverse scenario: - Peaking at 37 percent in 2026, 25 percentage points higher than in the baseline projection. - Investment valuation impact: - Decline of 7 percent in year 1 due to market and credit risk shocks. - Profitability: - One insurer would record a small operating loss in year 1 (before becoming profitable again in year 2); the two other entities remain profitable in each year. - Mortgage insurers might consider cutting or suspending dividends to maintain capital. - Available capital and solvency coverage: - Available capital would increase by 19 percent in the baseline scenario and by 6 percent in the adverse scenario in the first two years (modeling assumes no dividends paid). - In the adverse scenario, solvency coverage increases to 193 percent in year 2, 21 percentage points lower than in the baseline scenario. - Year 3: 217 percent in the adverse versus 231 percent in the baseline. - Result summary (concluded): - In the adverse scenario, combined ratio would peak at 37 percent in year 2, three times higher than in the baseline, but still indicating underwriting profits. - Available capital increases even in the adverse scenario but would be around 11 percent lower than in the baseline in year 2. - Mortgage insurers are largely resilient to market risk shocks, but investment valuations would still decline by around 7 percent in the first year. - Solvency ratios rise by 10 percentage points over two years in the adverse scenario, but by 32 percent in the baseline. - Source of calculations: IMF staff calculations based on OSFI data and company disclosures. ### Pensions’ funding risk analysis — scope and methodology - Scope: - Top-down funding risk analysis covered 40 DB pension plans supervised by OSFI and FSRA. - Sample split: 7 Ontario-chartered large public sector pension plans with average assets of CAD 84 billion; 33 private sector pension plans with average assets of CAD 8 billion and the largest private plan managing CAD 31 billion. - Scenario: - Used the same baseline and adverse scenario as the insurance risk analysis with minor adjustments for pension reporting granularity (e.g., specific shocks for investment fund holdings that replicate shocks for underlying asset classes). - Methodology and assumptions: - Projected valuation changes of assets and liabilities and funding ratio over a 3-year horizon. - Shocks applied to investment assets and liabilities as of June 2024 and subsequently for June 2025 and June 2026. - Haircuts in line with the adverse scenario applied to market value of investments. - Liabilities after stress approximated through a duration-based approach. - For hybrid plans, valuation change of DC assets replicated in valuation of DC liabilities. - Pension plans assumed to maintain asset allocation over full projection horizon and re-balance annually. - Contributions and benefit payments assumed to change in line with projected consumer prices in the scenario. - Funding ratio measured on a solvency basis. - Data sources: Regulatory returns provided by OSFI and FSRA, including annual financial statements, annual information returns, investment summaries, and most recent actuarial valuation reports. ### Pension scenario analysis — results and drivers - Starting position: Pension sector generally starts from a robust funding position with only five plans having a funding deficit prior to simulated stress. - Public sector plans (median plan): - Baseline: investment assets increase at rates between 3 and 4 percent per year; liabilities rise between 0.5 and 1.5 percent annually. - Adverse scenario: - Asset values decline by 11 percent in the first year; liability values decline by 13 percent in year 1. - Year 2: assets decline by 9 percent; liabilities increase by 6 percent. - Year 3: assets increase by 9 percent; liabilities increase by 15 percent. - Private sector plans: - Valuation changes are more varied across plans but directions and overall impacts comparable to public sector plans. - For the median private sector plan, changes in liability values in the adverse scenario are almost identical to public sector plans. - Contribution of shocks to funding surplus: - Interest rate shock is the major contributor to funding surplus change, dominating years 1 and 3. - Real estate shock impact larger than the equity shock for public sector plans. - Private sector plans show sizable contributions from the equity shock, indicating less diversification. - Depreciation of the CAD in years 1 and 2 is beneficial because plans typically hold sizable net FX positions through investments. - Funding ratio dynamics: - Adverse scenario year 1: offset of negative and positive shocks leads most public sector plans to see funding ratio increases (median public plan declines slightly); median private plan funding ratio increases by 4 percentage points. - Year 2 adverse scenario: funding ratios fall by 13 percentage points for the median large public sector plan and by 16 percentage points for the median private sector plan. - Year 3: funding ratios stabilize and increase slightly for most plans. - By end of year 3, a bit more than a quarter of plans would experience a funding deficit (which would not trigger an immediate refunding requirement). - Funding ratio levels (median plans): - Baseline, large public sector plan: from 128 to 140 percent in year 3. - Baseline, median private sector plan: from 118 to 132 percent over 3-year horizon. - Adverse scenario, median large public sector plan: funding ratios decline and reach 112 percent in year 3. - Adverse scenario, median private sector plan: reach 107 percent in year 3. ### Pension sensitivity analysis: equities and FX - Instantaneous 40 percent decline in equity prices would lower funding ratios: - Median public plan: decline by 4 percentage points. - Median private plan: decline by 10 percentage points. - Even among large public sector plans, a few would see a funding ratio below 100 percent. - A 10 percent appreciation of the CAD: - Median public sector pension plan funding ratio would decline by 3 percentage points. - Median private sector pension plan funding ratio would decline by 6 percentage points. - Source of calculations: IMF staff calculations based on OSFI and FSRA data. *Source: IMF staff calculations based on OSFI, AMF, and FSRA data as presented in the provided chapter.* ### 113. Through a bottom-up data collection, the FSAP tested the liquidity risks of pension ### 1canea2025008-source-pdf - 113. Through a bottom-up data collection, the FSAP tested the liquidity risks of pension ### Sample and scope - Nineteen pension plans covered with total assets of around CAD 1.83 trillion, about 70 percent of the Canadian pension sector. - Sub-sample composition: - Ten public sector pension plans (including a pension fund), registered in Ontario and British Columbia, as well as one plan under joint federal-provincial supervision, and one plan overseen at the federal level by the Treasury Secretariat Board. - Nine federally registered private sector pension plans, supervised by OSFI. - Three large life insurers supervised by OSFI were included; their total balance sheet assets amount to around CAD 1.70 trillion, representing about 85 percent of the total Canadian life insurance sector. ### Scenario and calibration - Historic benchmark: second half of September 1992 episode; reference date of the exercise: 30 June 2024. - Shock assumptions: - An increase of short-term interest term rates (up to 2 years) by 150 basis points, and long-term rates (10 years and longer) by 50 basis points (applicable to interest rates in all currencies); for maturities between 2 and 10 years, interest rate shocks were interpolated. - A depreciation of the CAD against the US dollar of 2.9 percent, and against the EUR of 4.8 percent. - Accordingly, the US dollar depreciates by 1.8 percent against the EUR. ### Methodology and modeling assumptions - Analysis compares liquidity need for margin and collateral calls against the stock of highly liquid high-quality assets and availability of other funding sources. - Timing assumptions: - After the shock at t=0, margin and collateral calls become due according to contractual arrangements: typically within one day (‘t+1’) or two days (‘t+2’). Remaining settlement periods are ‘t+3 or later’. - Participants reported: - The stock of available liquid assets (cash, deposits, different types of assets liquidatable within the projection horizon). - The flow of funding sources used on each of the three days, differentiating between (i) use of cash and deposits, (ii) use of liquid assets, and (iii) financing transactions such as credit lines and repurchase agreements. - Assumption: no commercial paper could be issued. ### Results — margin and collateral calls - Aggregate margin calls in the tested scenario: CAD 32 billion: - Pensions sector: CAD 26.4 billion (of which CAD 24.0 billion due within t+1). - Life insurers: CAD 5.5 billion (of which CAD 3.8 billion related to interest rate derivatives and mostly due within t+2). - Timing: Around 81 percent of the margin and collateral calls would need to be met within one day. - Margin call composition: - 47 percent from interest rate derivatives. - 21 percent from currency derivatives. - 32 percent from other off-balance sheet positions. - Among life insurers, most margin calls relate to interest rate derivatives. ### Results — liquid assets composition and valuation impact - Average shares of total assets held in liquid investments: - Sampled pension plans: 35 percent. - Three life insurers: 23 percent. - Composition of liquid assets (pre-stress): - Pensions sector: listed equities 45 percent, highly-rated government bonds 33 percent (dominant). - Life insurers: highly-rated corporate bonds 33 percent, government bonds 26 percent, BBB-rated bonds and money-market instruments 21 percent (broader mix). - Valuation impact of the scenario on liquid assets: - On average, the value declines by 2.2 percent. - Equity exposures rise in value by 1.9 percent and deposits by 0.9 percent (due to CAD depreciation). - Heterogeneity and coverage: - For half of the sample, liquid assets range between 31 and 64 percent of total assets. - For half of the sample, liquid assets exceed the simulated margin calls by a factor between 14 and 32. - Typical valuation impact of liquid assets in the scenario: between -1.7 and -4.2 percent. ### Results — sources of liquidity and settlement types - Aggregate sourced liquidity exceeds margin calls: - Sourced liquidity (aggregate): CAD 45 billion versus CAD 32 billion margin calls—driven by a few entities; most entities matched sourced liquidity closer to needed liquidity. - Pension plans: - Would source CAD 39 billion to meet margin calls. - Composition of sourced liquidity: - 55 percent from financing transactions (repos and credit lines). - 31 percent from highly liquid assets (mostly posting these instead of cash as collateral). - Remaining from cash and deposits. - Financing transactions composition: - Expiring repurchase agreements: 39 percent. - Uncommitted repos: 35 percent. - Committed credit lines: 20 percent. - Liquid assets used to meet margin calls: primarily government bonds and money-market instruments in rating categories AAA to A (85 and 15 percent, respectively). - Life insurers: - Would rely mostly on liquid asset holdings: 91 percent of sourced liquidity. - Liquid assets used: highly rated government bonds 74 percent and corporate bonds 25 percent. - Settlement in kind versus cash: - 53 percent of aggregated margin and collateral calls can be settled in kind. - By derivative type: - Interest-rate derivatives: in-kind settlement accounts for only 28 percent (cash settlement dominant). - FX derivatives: 54 percent in-kind. - Other off-balance sheet positions: 89 percent in-kind. - Life insurers: able to settle almost all their margin calls in kind. ### Recommendations - Insurers sector: - Adopt a comprehensive approach for macroprudential stress tests in the insurance sector. - Supervisory authorities should use stress tests to analyze systemic risks, cooperating in the design of the exercise and the scenario (also in consultation with the BOC). - Authorities should analyze potential second-round effects, including those stemming from insurers’ management actions, for example related to investment re-allocations. - Pension sector supervision and data: - OSFI and FSRA should further strengthen risk analyses. - Require more frequent and granular data on the largest pension plans’ investments, derivative holdings, and key metrics on liquidity and leverage to allow top-down sensitivity testing in times of heightened market volatility (most data in the pension sector are currently reported only on an annual basis). - Liquidity monitoring: - Extend monitoring of LCRs (already performed by FSRA for large public sector plans) to other large pension plans (including those supervised by other provincial authorities and OSFI) and to plans that are very active users of derivatives and other off-balance sheet transactions. - Close monitoring of systemic liquidity risks (together with leverage) should also cover pension funds under joint federal-provincial oversight and oversight by the Treasury Board Secretariat. *Source: IMF staff calculations based on company submissions and disclosures.* ### Appendix II. Macroeconomic Scenarios ### Appendix II. Macroeconomic Scenarios ### U.S. and Canada macro paths (figure highlights) - Figure shows comparative paths for baseline and adverse scenarios for 2024–2027 for: - House Prices Change YoY (CANADA, 2024-2027) - Stock Price YoY Change (USA and CANADA, 2024-2027) - Short Term Rate (USA, 2024-2027) - Real GDP Growth YoY (USA, 2024-2027) - Unemployment (USA, 2024-2027) - Sources: World Economic Outlook and Staff Calculations. ### Banking Sector: Solvency Test — Institutional perimeter - Institutions included: - Seven systemic deposit taking institutions (DTIs), including Desjardins and six D-SIBs: Royal Bank of Canada, Toronto-Dominion Bank, Bank of Nova Scotia, Bank of Montreal, Canadian Imperial Bank of Commerce, and National Bank of Canada. - Royal Bank of Canada and Toronto-Dominion Bank are also considered global systemically important banks (G-SIBs). - Market share: - The six D-SIBs represent about 93.6 percent of banking sector assets (excluding foreign bank branches). - The seven systemic DTIs represent above 90.2 percent of deposit-taking institutions assets. - Data and baseline date: - Data sources include OSFI regulatory returns and supervisory data, loan level information from the RESL dataset; AMF regulatory returns and loan level information for RESL. - Scope of consolidation: Global consolidated group basis. For Desjardins insurance business activities are excluded. - Cut-off date: end-October 2024. ### Banking Sector: Solvency Test — Channels of risk propagation and methodology - Methodology: - Balance-sheet based approach. - Projections of key balance sheet, income statement, and capital account items conditional on scenarios. - Static balance sheet assumption. - Satellite models for macro-financial linkages: - Credit Risk: - Mortgage Loans: household financial conditions adjusted to macro developments; employment dynamics modeled; loan-level risks assessed through iterative simulations; PD estimated using a BMA approach. (Refer to Household Vulnerability Analysis STeM section.) - Corporate Loans: corporate stress test satellite models link credit risk variables with macro variables estimated using a BMA methodology. - Net Interest Income: - Projected using a structural model reflecting repricing dynamics of banks’ balance sheets. - Historical data on interest rates for newly originated interest-earning assets and liabilities used to estimate pass-through of policy rates. - Non-performing loans will not generate any income. - Market risk: - Valuation losses from full revaluation of sovereign securities, corporate fixed income debt securities and equity holdings using a Mark to Market (MTM) approach for fair-valued securities. - Market risk estimated as a sensitivity analysis. - Other P&L components: - Econometric models estimated for fees and commission income and other income/expenses. - Stress test horizon: - 3 years (2025–27). ### Banking Sector: Solvency Test — Tail shocks / Scenario analysis - Two macroeconomic scenarios: - Baseline scenario based on the October 2024 WEO projections. - Adverse scenario consistent with the FSAP RAM featuring deepening geoeconomic fragmentation: - Greater protectionism and increased use of cross-border restrictions (trade, migration, FDI, technology diffusion). - Impediments to global trade via non-tariff and tariff barriers, including “trade wars” between some regions; sharp de-integration of North American supply chains. - Persistent supply shock, downward pressure on global growth, temporary supply shortages, increased inflation expectations. - Central banks pause cutting interest rates (or reverse recent cuts). - Slower global growth increases unemployment and reduces commodity demand, causing oil prices to drop. - Reduced FDI and technological diffusion dent innovation and productivity. - Widespread risk-off event and asset valuation corrections. - Elevated borrowing rates, higher unemployment, reduced household income produce sharp corrections in residential real estate and CRE. ### Banking Sector: Solvency Test — Risks and buffers; behavioral assumptions - Positions/risk factors assessed: - Credit risk (provision costs): - Estimated according to Basel III framework. - Captures all on-balance sheet exposures at amortized cost by sector with different paths per sector. - Starting credit parameters used to project scenario conditional forward paths. - Sovereign risk: - Mark-to-market valuation of sovereign securities from shocks to interest rates and credit spreads linked to macro scenario. - Market risk other than sovereign: - Valuation effects of FVTPL and FVOCI positions. - Profits: - Net interest income affected by change in reference rate and pass-through. - Net fee and commission income and other income/expense evolve with macro conditions and banks’ balance sheets. - Behavioral adjustments: - Balance sheet composition remains constant over the stress test horizon. - No recognized interest on non-performing exposures. - Maturing assets are replaced by exposures of the same type and risk. - Statutory tax rates applied. - DTIs can only accumulate capital through retained earnings. - If DTIs’ capital ratio falls below regulatory minimum during the stress test horizon, no prompt corrective action is assumed. - Dividends linked to DTIs’ net profits: - Under positive profits, and if the Capital Conservation Buffer (CCoB) is not breached, the dividend payout ratio is set at the ratio observed at the cutoff point (T0). - If the CCoB is breached, restrictions on dividend distributions are aligned with the regulatory framework. ### Banking Sector: Solvency Test — Regulatory and market-based standards - Calibration of risk parameters: - Scenario dependent forward paths for Point-in-Time (PiT) PDs estimated for each type of exposure. - For IRB exposures, RWA projected on basis of updated regulatory through-the-cycle PDs and downturn LGDs, using appropriate scaling multipliers from the PiT parameters. - For standardized approach (STA) exposures, RWA projected based on constant risk weight densities. - Regulatory standards: - Baseline hurdles include regulatory minimum, the CCoB, the D-SIBs surcharge and the Domestic Stability Buffer (DSB). The DSB is not required for Desjardins. - In the adverse scenario, DTIs are allowed to deplete the CCoB and the DSB. Other requirements remain in place. - Hurdle rates are based on common equity tier-1, tier-1, and total capital ratios. ### Banking Sector: Solvency Test — Reporting format for results - Output presentation: - Evolution of CET1 for the seven systemic DTIs in aggregate. - Decomposition of key drivers to aggregate net profits and aggregate CET1 capital ratios, including differences between baseline and adverse scenarios. - Number of DTIs and share of total assets below hurdle rates. ### Banking Sector: Liquidity Test — Institutional perimeter and data - Institutions included: - Seven systemic DTIs, including Desjardins and six D-SIBs: Royal Bank of Canada, Toronto-Dominion Bank, Bank of Nova Scotia, Bank of Montreal, Canadian Imperial Bank of Commerce, and National Bank of Canada. - Royal Bank of Canada and Toronto-Dominion Bank considered G-SIBs. - Market share: - The six D-SIBs represent about 93.6 percent of banking sector assets (excluding foreign bank branches). - The seven systemic DTIs represent above 90.2 percent of deposit-taking institutions assets. - Data and baseline date: - OSFI: Regulatory returns based on the Liquidity Coverage Ratio (LCR) and Net Cumulative Cash Flow (NCCF) for the 6 D-SIBs. - AMF: Regulatory returns based on the LCR and the NCCF for Desjardins. - Scope of consolidation: Global consolidated group basis. - Cut-off date: end-October 2024. ### Banking Sector: Liquidity Test — Channels, methodology, and horizon - Methodology: - Two types of tests: LCR test and cash-flow analysis. - LCR test: - In line with standard Basel monitoring tool, featuring total liquidity and liquidity in significant currencies (Canadian dollar, U.S. dollar, euro, British pound, and Japanese yen). - Scenarios for LCR outflows and HQLA haircuts produce stressed LCR ratios. - Stress test horizon: 30 days. - Cash-flow analysis: - Assesses DTIs' liquidity risk using cumulative net funding gap and counterbalancing capacity. - Net funding gap: difference between cash inflows and outflows in each time bucket, summed across buckets within horizon. - Counterbalancing capacity: cumulative value of liquid assets that DTIs can liquidate under stress at reasonable prices (mainly cash resources and securities). - Liquidity shortfall arises when counterbalancing capacity insufficient to meet net funding gap. ### Banking Sector: Liquidity Test — Risks, behavioral adjustments, scenarios, and standards - Risks: - Funding liquidity risk reflected in funding run-off rates. - Market liquidity risk reflected in asset haircuts, influenced by market movements, potential fire sales and collateral supply. - Behavioral adjustments: - Liquidity from the Bank of Canada’s Emergency Lending Assistance (ELA) is not considered. - Tail shocks / Scenario analysis: - LCR test: various scenarios with varying intensity of adverse liquidity conditions. - Cash-flow analysis: 20 scenarios from mild to severely adverse liquidity conditions; considers both funding and market liquidity risks. - Regulatory standards and parameters: - LCR hurdle rate set at 100 percent at the aggregate currency level (per Basel III). ### Banking Sector: Liquidity Test — Reporting format for results - Output presentation: - Changes in system-wide liquidity position, including drivers for cash outflows, inflows, and counterbalancing capacity. - Distribution of DTIs’ liquidity positions. - Number of institutions with LCR below 100 percent and/or negative net cash balance. - Amount of liquidity shortfalls. ### Interconnectedness and Contagion Analysis - Institutions involved: - Confidential interbank exposure data: six D-SIBs and aggregate data for remaining banks (if available). - Publicly available cross-sectoral and cross-border data: Statistics Canada financial accounts on a from-whom-to-whom basis. - Data and starting position: - Supervisory data: 2024: Q2 (or most recently available). - Methodology: - Contagion and interconnectedness Network Analysis: EVS Model, 2010. - Risks: - Credit and funding losses related to bilateral exposures, and fire-sale of assets following sizeable withdrawals of deposits. - Cross-border exposures (data permitting). - Buffers: - Institution’s own capital and liquidity buffers, sector’s aggregate capital buffers. - Size of shocks: - Default of institutions. - Output/Presentation: - Network charts: Economy-wide and inter-financial network based on exposures. - Entity-level contagion/vulnerability/amplification indices. ### Household Sector Vulnerability Analysis - Objective: - Assess overall indebtedness in the household sector under FSAP baseline and adverse scenario. Project mortgage PD as input to banking sector stress test. - Data: - OSFI mortgage loan-level RESL data, September 2023 version, supplemented by aggregated statistics from Statistics Canada. - Bank-level historical mortgage PD path shared by BOC. - Methodology: - Two-stage approach integrating structural simulation with a Bayesian econometric framework. - Stage 1 — Structural Simulation: - (1) Baseline Initialization: Update household financial variables to reflect end-2024 conditions. - (2) Employment Status Simulation: Bootstrap approach to model employment evolution aligning with projected unemployment trends. - (3) Income and debt Metrics Update: Adjust income based on employment status and wage growth, feeding into DSR. - (4) Aggregation: Repeat stochastic process across iterations, aggregating results into a debt-weighted PD estimate for projection horizon. - Stage 2 — BMA Approach: - (1) DSR Anchoring: Aligns projected debt service ratios with end-2024 aggregate statistics from Statistics Canada. - (2) Econometric Estimation: Uses a by-bank panel BMA framework to refine PD projections by selecting relevant macro-financial predictors. - FSAP Systematic Risk workstream collaborates with Bank of Canada to refine HRAM scenario and leverage BOC access to household survey data, credit bureau records and tax survey data that FSAP team does not have access to obtain estimates of BOC’s PD path as reference. ### Corporate Sector Vulnerability Analysis - Objective: - Quantify share of financially weak NFCs and assess corporate sector resilience. - Project PD at aggregate and industry level as input to banking sector stress test. - Data: - Firm-level balance sheet, income statement and credit data from Moody’s Orbis and Moody’s KMV EDF. - Statistics Canada National Balance Sheet Account for aggregate NFC vulnerability indicator construction. - Overall and industry-level historical corporate PD shared by BOC. - Methodology: - Use Statistics Canada NBSA for headline corporate vulnerability indicators (debt-to-asset ratio, cash-to-debt ratio, debt-to-equity ratio). - Integrate Moody’s Orbis firm-level data and Moody’s KMV 1-year EDF focusing on leverage, profitability (EBIT Ratio, ROA, ROE), and liquidity (current ratio, interest coverage ratio, cash buffer) at aggregate and by firm-size. - Identify financially weak firms based on debt-servicing capacity; examine trends in firms-at-risk and debt-at-risk. - Conduct firm-level fixed effects regressions to analyze PD drivers following IMF Japan FSAP (2024) approach. - Project PDs using historical corporate PDs provided by BOC and macro scenarios, applying BMA methods from Gross and Población (2019) at aggregate and industry levels. - Incorporate FactSet trends on publicly traded NFCs provided by authorities. - Address SME data limitations using supplementary information provided by authorities. ### Pension Funds: Solvency Stress Test (Top-down) - Institutional perimeter: - Number of institutions: - 7 Public sector pension plans (Ontario) - 33 Private sector single-employer and multi-employer pension plans (federal, Ontario) - Market share: - ~35 percent, based on assets - Data: - Statutory returns - Reference date: - June 30, 2024 - Channels of risk propagation and methodology: - Investment assets: market value changes of assets after price shocks. - Liabilities: valuation change due to interest rate shock. - Impact measured on net assets (difference between stressed assets and liabilities) and funding ratios. - Time horizon: 3 years. - Scenario analysis — Adverse scenario aligned with macrofinancial scenario with granularity: - Canadian stocks: -17.3 percent (year 1), -18.1 percent (year 2), +14.3 percent (year 3) - Canadian commercial real estate: -14.0 percent (year 1), -5.4 percent (year 2), +7.9 percent (year 3) - Canadian short-term risk-free interest rates: -33 bps (year 1), -80 bps (year 2), -22 bps (year 3) - Canadian short-term risk-free interest rates (alternate series): +114 bps (year 1), -53 bps (year 2), -133 bps (year 3) - Canadian sovereign bond spreads: +25 bps (year 1), +30 bps (year 2), -39 bps (year 3) - Corporate bond spreads: - year 1: between +20 bps for AAA-rated and +101 bps for BB and lower - year 2: between +25 bps for AAA-rated and +124 bps for BB and lower - year 3: between -32 bps for AAA-rated and -158 bps for BB and lower - Canadian dollar (external value): -4.7 percent (year 1), -4.0 percent (year 2), +3.1 percent (year 3) - Sensitivity analysis (instantaneous market risk shocks): - Stock price decline: -40 percent for ordinary shares, -20 percent for preferred shares - CAD appreciation: +10 percent - Risk factors assessed: - Market risks (equity, property, FX, alternative assets) - Interest rate risks - Credit risks (sovereign and corporate bond spreads) - Regulatory/accounting standards: - Canadian Accounting Standards for Pension Plans (ASPP) - Reporting formats for results: - Change in values of assets and liabilities - Funding ratios - Contribution of individual shocks ### Pension Funds: Liquidity Risk (Bottom-up) - Institutional perimeter: - Number of institutions: - 10 Public sector pension plans (Ontario, British Columbia, federal, jointly federal-provincial) - 9 Private sector single-employer pension plans (federal) - Market share: - ~70 percent, based on assets - Data: - Data collection from participating pension plans - Reference Date: - 30 June 2024 *International Monetary Fund — Appendix II. Macroeconomic Scenarios (extracted content).* ### 8. Channels of Risk ### 8. Channels of Risk ### Overview - Focus: Channels through which market, interest rate, FX, credit, and liquidity shocks propagate to insurers and pension plans over short and medium horizons. - Tests covered: Combination of margin-call scenarios for pension plans and insurers; Insurers Solvency Stress Test (Top-down); Insurers Liquidity Stress Test (Bottom-up); Pension plan ASPP scenario. ### Pension plans — ASPP margin-call test - Methodology: - Combination of interest rate and FX shocks leading to margin calls on pension plans’ derivative and other off-balance sheet positions. - Time horizon: - Up to three days - Scenario (Tail shocks): - Instantaneous increase of interest rates (short-term +150 basis points, long-term +50 basis points; for all currencies) - CAD depreciation: -2.9 percent against USD, and -4.8 percent against EUR - Risk factors assessed: - Short-term liquidity risks - Regulatory/accounting standards: - Canadian ASPP - Reporting formats / Output presentation: - Amount of margin call (per day) - Share of margin calls which could be met in kind - Liquid assets - Sources of liquidity to meet margin calls ### Insurers Solvency Stress Test — Top-down 1. Institutional perimeter - Number of institutions: - ~9 life insurers - ~17 P&C insurers - 3 mortgage insurers - Market share: - Life: ~94 percent, based on balance sheet assets - P&C: ~57 percent, based on gross premiums - Mortgage: 100 percent, based on gross premiums - Consolidation level: Unconsolidated - Data: Statutory returns - Reference Date: June 30, 2024 2. Channels of risk propagation / Methodology - Investment assets: market value changes of assets after price shocks - Liabilities: valuation change due to interest rate shock - Impact on available capital: net assets as the difference between stressed assets and liabilities - For mortgage insurers: higher claims through heightened defaults on insured mortgage loans - Time horizon: 3 years 3. Scenario analysis (Tail shocks / Adverse scenario) - Canadian stocks: -17.3 percent (year 1), -18.1 percent (year 2), +14.3 percent (year 3) - Canadian commercial real estate: -14.0 percent (year 1), -15.4 percent (year 2), +7.9 percent (year 3) - Canadian short-term risk-free interest rates: -33 bps (year 1), -80 bps (year 2), -22 bps (year 3) - [Duplicate line in source] Canadian short-term risk-free interest rates: +114 bps (year 1), -53 bps (year 2), -133 bps (year 3) - Canadian sovereign bond spreads: +25 bps (year 1), +30 bps (year 2), -39 bps (year 3) - Corporate bond spreads: - year 1: between +20 bps for AAA-rated and +101 bps for BB and lower - year 2: between +25 bps for AAA-rated and +124 bps for BB and lower - year 3: between -32 bps for AAA-rated and -158 bps for BB and lower - Canadian dollar (external value): -4.7 percent (year 1), -4.0 percent (year 2), +3.1 percent (year 3) 4. Sensitivity analysis — Instantaneous market risk shocks - Stock price decline: - -40 percent for ordinary shares - -20 percent for preferred shares - CAD appreciation: +10 percent 5. Risk factors assessed - Market risks (equity, property, FX, alternative assets) - Interest rate risks - Credit risks (sovereign and corporate bond spreads) - For mortgage insurers: credit risks from underwriting business 6. Regulatory/accounting standards - IFRS 17 7. Reporting formats / Output presentation - Change in valuation of assets and liabilities - Solvency ratios - Aggregated capital shortfall (where applicable) - Dispersion across companies - Contribution of individual shocks ### Insurers Liquidity Stress Test — Bottom-up 1. Institutional perimeter - Number of institutions: - 3 life insurers - Market share: - Life: ~85 percent, based on balance sheet assets - Consolidation level: Unconsolidated - Data: Data collection from participating life insurers - Reference date: June 30, 2024 2. Channels of risk propagation / Methodology - Combination of interest rate and FX shocks leading to margin calls on insurers’ derivative and other off-balance sheet positions - Time horizon: Up to three days 3. Scenario analysis (Tail shocks) - Instantaneous increase of interest rates (short-term +150 basis points, long-term +50 basis points; for all currencies) - CAD depreciation: -2.9 percent against USD, and -4.8 percent against EUR 4. Risk factors assessed - Short-term liquidity risks 5. Regulatory/accounting standards - IFRS 17 6. Reporting formats / Output presentation - Amount of margin call (per day) - Share of margin calls which could be met in kind - Liquid assets - Sources of liquidity to meet margin calls ### Appendix highlights (Selected methodological details) - Structural model for interest income (DTI-segment level): - Inputs: - Repricing ladder at T0: exposures by repricing bucket, 푃푃0[k,k+1], with fraction 휃휃0[k,k+1] - Scenario-specific projections for the interest rate on new originations, denoted as 푝푝푡푡NNNN - Key modeling steps: - Step 1: Simulate exposures originated/repriced in each bucket and period using the law of motion: - Equation (1): 푃푃푡푡[푘푘−1,푘푘] = 푃푃푡푡−1[푘푘,푘푘+1] + 퐼퐼푡푡[푘푘−1,푘푘] for 푘푘=1,2,3 - Assumption: shares of exposures across buckets are constant over time: - Equation (2): 휃휃푡푡[푘푘,푘푘+1] = 휃휃0[푘푘,푘푘+1] - Step 2: Simulate average interest rate for each bucket and period: - Equation (3): 푝푝푡푡[푘푘−1,푘푘] = 훿훿푡푡 푝푝푡푡−1[푘푘,푘푘+1] + (1−훿훿푡푡) 푝푝푡푡NNNN, where 훿훿푡푡 = 푃푃푡푡−1[푘푘,푘푘+1] / 푃푃푡푡[푘푘−1,푘푘] - Initial condition: initial interest rate in all buckets equals average interest rate of the portfolio at T0 - Step 3: Calculate interest income: - Equation (4): 퐼퐼퐼퐼푡푡 = ∑_{푘푘=1}^{3} 푝푝푡푡−1[푘푘,푘푘+1] 푃푃푡푡−1[푘푘,푘푘+1] + 1/2 푝푝푡푡−1[0,1] 푃푃푡푡−1[0,1] + 1/2 푝푝푡푡NNNN ∑_{푘푘=0}^{3} 퐼퐼푡푡[푘푘,푘푘+1] - Timing assumptions: exposures with <1 year to repricing and newly issued/repriced loans enter/priced at mid-year (earning half-year rates) ### Liquidity Coverage Ratio-based test scenarios — Haircuts and Run-off Rates (selected entries) - Haircuts (examples): - Coins and banknotes: 1.00, 1.00, 1.00 - Total central bank reserves; part drawable in stress: 1.00, 1.00, 1.00 - Securities with a 0% risk weight (issued by sovereigns): 1.00, 0.95, 0.90 - Securities with a 20% risk weight (issued by sovereigns): 0.85, 0.80, 0.75 - Non-financial corporate bonds, rated AA- or better: 0.85, 0.75, 0.65 - Residential mortgage-backed securities (RMBS), rated AA or better: 0.75, 0.65, 0.55 - Non-financial common equity shares: 0.50, 0.40, 0.30 - Run-off rates (selected retail / wholesale entries): - Total retail deposits eligible for a 3% run-off rate in Canada: 0.03, 0.10, 0.03, 0.10 (layout in source presents multiple cells) - Less stable retail deposits — uninsured deposits: 0.10, 0.20, 0.10, 0.20 - Total unsecured wholesale funding — less stable deposits in non-transactional and non-relationship accounts: 0.10, 0.10, 0.30, 0.30 - Term deposits managed by an unaffiliated third-party — cashable or maturing in next 30 days: 0.30, 0.30, 0.50, 0.50 - Demand deposits managed by unaffiliated third-party: 0.40, 0.40, 0.65, 0.65 *Source: IMF staff.* ### 2. Net cash outflows ### 2. Net cash outflows ### 2.1. Cash outflows - 2.1.1. Retail deposit run-off - Term deposits with a remaining maturity of >30 days: 0.00 0.00 0.00 - Total operational deposits; of which: - provided by non-financial corporates - insured, with a 3% run-off rate: 0.03 0.03 0.15 0.15 - insured, with a 5% run-off rate: 0.05 0.03 0.20 0.20 - uninsured: 0.25 0.25 0.40 0.40 - provided by sovereigns, central banks, PSEs and MDBs - insured, with a 3% run-off rate: 0.03 0.03 0.15 0.15 - insured, with a 5% run-off rate: 0.05 0.03 0.20 0.20 - uninsured: 0.25 0.25 0.40 0.40 - provided by banks - insured, with a 3% run-off rate: 0.03 0.03 0.15 0.15 - insured, with a 5% run-off rate: 0.05 0.03 0.20 0.20 - uninsured: 0.25 0.25 0.40 0.40 - provided by other financial institutions and other legal entities - insured, with a 3% run-off rate: 0.03 0.03 0.15 0.15 - insured, with a 5% run-off rate: 0.05 0.03 0.20 0.20 - uninsured: 0.25 0.25 0.45 0.45 - Total non-operational deposits; of which - provided by non-financial corporates; of which: - where entire amount is fully covered by an effective deposit insurance scheme: 0.20 0.20 0.40 0.40 - where entire amount is not fully covered by an effective deposit insurance scheme: 0.40 0.40 0.60 0.60 - provided by sovereigns, central banks, PSEs and MDBs; of which: - where entire amount is fully covered by an effective deposit insurance scheme: 0.20 0.20 0.40 0.40 - where entire amount is not fully covered by an effective deposit insurance scheme: 0.40 0.40 0.60 0.60 - provided by other banks: 1.00 1.00 1.00 1.00 - provided by other financial institutions and other legal entities: 1.00 1.00 1.00 1.00 - 2.1.2. Unsecured wholesale funding run-off - Unsecured debt issuance: 1.00 1.00 1.00 1.00 - Additional balances required to be installed in central bank reserves: 1.00 1.00 1.00 1.00 - Total unsecured wholesale funding run-off: (implied by line items above) - Transactions maturing ≤ 30 days conducted with the bank's domestic central bank; of which: - Backed by Level 1 assets; of which: - Transactions involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Backed by Level 2A assets; of which: 0.00 0.00 0.00 0.00 - Transactions involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Backed by Level 2B RMBS assets; of which: 0.00 0.00 0.00 0.00 - Transactions involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Backed by Level 2B non-RMBS assets; of which: 0.00 0.00 0.00 0.00 - Transactions involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Backed by other assets: 0.00 0.00 0.00 0.00 - Transactions maturing ≤ 30 days not conducted with the bank's domestic central bank and backed by Level 1 assets; of which: - Transactions involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions maturing ≤ 30 days not conducted with the bank's domestic central bank and backed by Level 2A assets; of which: - Transactions involving eligible liquid assets: 0.15 0.15 0.15 0.15 - Transactions not involving eligible liquid assets: 0.15 0.15 0.15 0.15 - Transactions maturing ≤ 30 days not conducted with the bank's domestic central bank and backed by Level 2B non-RMBS assets; of which: - Transactions involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Transactions not involving eligible liquid assets: 0.25 0.25 0.25 0.25 ### 2.1.3. Secured funding run-off - Transactions maturing ≤ 30 days not conducted with the bank's domestic central bank and backed by Level 2B non-RMBS assets; of which: - Counterparties are domestic sovereigns, MDBs or domestic PSEs with a 20% risk weight; of which: - Transactions involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Transactions not involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Counterparties are not domestic sovereigns, MDBs or domestic PSEs with a 20% risk weight; of which: - Transactions involving eligible liquid assets: 0.50 0.50 0.50 0.50 - Transactions not involving eligible liquid assets: 0.50 0.50 0.50 0.50 - Transactions maturing ≤ 30 days not conducted with the bank's domestic central bank and backed by other assets (non-HQLA); of which: - Counterparties are domestic sovereigns, MDBs or domestic PSEs with a 20% risk weight: 0.25 0.25 0.25 0.25 - Counterparties are not domestic sovereigns, MDBs or domestic PSEs with a 20% risk weight: 1.00 1.00 1.00 1.00 - Total secured wholesale funding run-off: (implied by line items above) - Derivatives cash outflow: 1.00 1.00 1.00 1.00 - Increased liquidity needs related to downgrade triggers in derivatives and other financing transactions: 1.00 1.00 1.00 1.00 - Increased liquidity needs related to the potential for valuation changes on posted collateral securing derivative and other transactions: - Cash and Level 1 assets: 0.00 0.00 0.00 0.00 - For other collateral (ie all non-Level 1 collateral): 0.20 0.20 0.20 0.20 - Increased liquidity needs related to excess non-segregated collateral held by the bank that could contractually be called at any time by the counterparty: 1.00 1.00 1.00 1.00 - Increased liquidity needs related to contractually required collateral on transactions for which the counterparty has not yet demanded the collateral be posted: 1.00 1.00 1.00 1.00 - Increased liquidity needs related to contracts that allow collateral substitution to non-HQLA assets: 1.00 1.00 1.00 1.00 - Increased liquidity needs related to market valuation changes on derivative or other transactions: 1.00 1.00 1.00 1.00 - Loss of funding on ABS and other structured financing instruments issued by the bank, excluding covered bonds: 1.00 1.00 1.00 1.00 - Loss of funding on covered bonds issued by the bank: 1.00 1.00 1.00 1.00 - Loss of funding on ABCP, conduits, SIVs and other such financing activities; of which: - debt maturing ≤ 30 days: 1.00 1.00 1.00 1.00 - with embedded options in financing arrangements: 1.00 1.00 1.00 1.00 - other potential loss of such funding: 1.00 1.00 1.00 1.00 - Undrawn committed credit and liquidity facilities to retail and small business customers: 0.05 0.05 0.05 0.05 - Undrawn committed credit facilities to: - non-financial corporates: 0.10 0.10 0.25 0.25 - sovereigns, central banks, PSEs and MDBs: 0.10 0.10 0.10 0.10 - Undrawn committed liquidity facilities to: - non-financial corporates: 0.30 0.30 0.50 0.50 - sovereigns, central banks, PSEs and MDBs: 0.30 0.30 0.30 0.30 - Undrawn committed credit and liquidity facilities provided to banks subject to prudential supervision: 0.40 0.40 0.40 0.40 - Undrawn committed credit facilities provided to other Fis: 0.40 0.40 0.40 0.40 - Undrawn committed liquidity facilities provided to other Fis: 1.00 1.00 1.00 1.00 - Undrawn committed credit and liquidity facilities to other legal entities: 1.00 1.00 1.00 1.00 - Other contractual obligations to extend funds to: - financial institutions: 1.00 1.00 1.00 1.00 - retail clients / small business customers / non-financial corporates / other clients: - retail, small business customers, non-financials and other clients: 1.00 1.00 1.00 1.00 - Total contractual obligations to extend funds in excess of 50% roll-over assumption: (implied by line items above) ### 2.1.4. Additional requirements - Other contingent funding obligations: - Non-contractual obligations related to potential liquidity draws from joint ventures or minority investments in entities: 1.00 1.00 1.00 1.00 - Unconditionally revocable "uncommitted" credit and liquidity facilities provided to retail and small business customers: 0.02 0.05 0.02 0.05 - Unconditionally revocable "uncommitted" credit and liquidity facilities provided to all other customers: 0.05 0.05 0.10 0.10 - Trade finance-related obligations (including guarantees and letters of credit): 0.03 0.03 0.03 0.03 - Guarantees and letters of credit unrelated to trade finance obligations: 0.05 0.10 0.05 0.10 - Non-contractual obligations: - Debt-buy back requests (incl related conduits): 0.00 0.00 0.00 0.00 - Structured products: 0.05 0.05 0.05 0.05 - Managed funds: 0.00 0.00 0.00 0.00 - Other non-contractual obligations: 0.05 0.05 0.05 0.05 - Outstanding debt securities with remaining maturity > 30 days: 0.00 0.00 0.00 0.00 - Non contractual obligations where customer short positions are covered by other customers’ collateral: 0.50 0.75 0.50 0.75 - Bank outright short positions covered by a collateralised securities financing transaction: 0.00 0.00 0.00 0.00 - Other contractual cash outflows (including those related to unsecured collateral borrowings and uncovered short positions): 1.00 1.00 1.00 1.00 - Total run-off on other contingent funding obligations: (implied by line items above) - Reverse repo and other secured lending or securities borrowing transactions maturing ≤ 30 days: - Of which collateral is not re-used (ie is not rehypothecated) to cover the reporting institution's outright short positions: - Transactions backed by Level 1 assets; of which: - Transactions involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Transactions backed by Level 2A assets; of which: - Transactions involving eligible liquid assets: 0.15 0.15 0.15 0.15 - Transactions not involving eligible liquid assets: 0.15 0.15 0.15 0.15 - Transactions backed by Level 2B RMBS assets; of which: - Transactions involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Transactions not involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Transactions backed by Level 2B non-RMBS assets; of which: - Transactions involving eligible liquid assets: 0.50 0.50 0.50 0.50 - Transactions not involving eligible liquid assets: 0.50 0.50 0.50 0.50 - Margin lending backed by non-Level 1 or non-Level 2 collateral: 0.50 0.50 0.50 0.50 - Transactions backed by other collateral: 1.00 1.00 1.00 1.00 - Of which collateral is re-used (ie is rehypothecated) in transactions to cover the reporting insitution's outright short positions: - Transactions backed by Level 1 assets: 0.00 0.00 0.00 0.00 - Transactions backed by Level 2A assets: 0.00 0.00 0.00 0.00 - Transactions backed by Level 2B RMBS assets: 0.00 0.00 0.00 0.00 - Transactions backed by Level 2B non-RMBS assets: 0.00 0.00 0.00 0.00 - Margin lending backed by non-Level 1 or non-Level 2 collateral: 0.00 0.00 0.00 0.00 - Transactions backed by other collateral: 0.00 0.00 0.00 0.00 - Total inflows on reverse repo and securities borrowing transactions: (implied by line items above) - Contractual inflows due in ≤ 30 days from fully performing loans, not reported as secured lending, from: - Retail customers: 0.50 0.50 0.50 0.50 - Small business customers: 0.50 0.50 0.50 0.50 - Non-financial corporates: 0.50 0.50 0.50 0.50 - Central banks: 1.00 1.00 1.00 1.00 ### 2.2. Cash inflows - Is reporting institution an indirect clearer that is not a subsidiary of an OSFI or provincially-regulated direct clearer? (question posed in source) - 2.2.1. Secured lending including reverse repo and securities borrowing - Financial institutions, of which: - operational deposits related to clearing activities placed by indirect clearers with an OSFI or provincially-regulated direct clearer: 0.25 0.25 0.25 0.25 - other operational deposits: 0.00 0.00 0.00 0.00 - all payments on other loans and deposits due in ≤ 30 days: 1.00 1.00 1.00 1.00 - Other entities: 0.50 0.50 0.50 0.50 - Total of other inflows by counterparty: (implied by line items above) - 2.2.3. Other cash inflows - Derivatives cash inflow: 1.00 1.00 1.00 1.00 - Contractual inflows from securities maturing ≤ 30 days, not included anywhere above: 1.00 1.00 1.00 1.00 - Other contractual cash inflows: 1.00 1.00 1.00 1.00 - Total of other cash inflows: (implied by line items above) - Total cash inflows before applying the cap: (implied by line items above) - Cap on cash inflows: 0.75 0.75 0.75 0.75 - Total cash inflows after applying the cap: (implied after cap) - Collateral swaps maturing ≤ 30 days: - Of which the borrowed assets are not re-used (ie are not rehypothecated) to cover short positions - Level 1 assets are lent and Level 1 assets are borrowed; of which: - Involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Level 2A assets are lent and Level 1 assets are borrowed; of which: - Involving eligible liquid assets: 0.15 0.15 0.15 0.15 - Not involving eligible liquid assets: 0.15 0.15 0.15 0.15 - Level 2A assets are lent and Level 2A assets are borrowed; of which: - Involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Level 2B RMBS assets are lent and Level 1 assets are borrowed; of which: - Involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Not involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Level 2B RMBS assets are lent and Level 2A assets are borrowed; of which: - Involving eligible liquid assets: 0.10 0.10 0.10 0.10 - Not involving eligible liquid assets: 0.10 0.10 0.10 0.10 - Level 2B RMBS assets are lent and Level 2B RMBS assets are borrowed; of which: - Involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Level 2B non-RMBS assets are lent and Level 2A assets are borrowed; of which: - Involving eligible liquid assets: 0.35 0.35 0.35 0.35 - Not involving eligible liquid assets: 0.35 0.35 0.35 0.35 - Level 2B non-RMBS assets are lent and Level 2B RMBS assets are borrowed; of which: - Involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Not involving eligible liquid assets: 0.25 0.25 0.25 0.25 - Level 2B non-RMBS assets are lent and Level 2B non-RMBS assets are borrowed; of which: - Involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Not involving eligible liquid assets: 0.00 0.00 0.00 0.00 - Other assets are lent and Level 1 assets are borrowed: 1.00 1.00 1.00 1.00 - Other assets are lent and Level 2A assets are borrowed: 0.85 0.85 0.85 0.85 - Other assets are lent and Level 2B RMBS assets are borrowed: 0.75 0.75 0.75 0.75 - Other assets are lent and Level 2B non-RMBS assets are borrowed: 0.50 0.50 0.50 0.50 - Other assets are lent and other assets are borrowed: 0.00 0.00 0.00 0.00 - Of which the borrowed assets are re-used (ie are rehypothecated) in transactions to cover short positions: - Level 1 assets are lent and Level 1 assets are borrowed: 0.00 0.00 0.00 0.00 - Level 2A assets are lent and Level 1 assets are borrowed: 0.15 0.15 0.15 0.15 - Level 2A assets are lent and Level 2A assets are borrowed: 0.00 0.00 0.00 0.00 - Level 2B RMBS assets are lent and Level 1 assets are borrowed: 0.25 0.25 0.25 0.25 - Level 2B RMBS assets are lent and Level 2A assets are borrowed: 0.10 0.10 0.10 0.10 - Level 2B RMBS assets are lent and Level 2B RMBS assets are borrowed: 0.00 0.00 0.00 0.00 - Level 2B non-RMBS assets are lent and Level 1 assets are borrowed: 0.50 0.50 0.50 0.50 - Level 2B non-RMBS assets are lent and Level 2A assets are borrowed: 0.35 0.35 0.35 0.35 - Level 2B non-RMBS assets are lent and Level 2B RMBS assets are borrowed: 0.25 0.25 0.25 0.25 - Level 2B non-RMBS assets are lent and Level 2B non-RMBS assets are borrowed: 0.00 0.00 0.00 0.00 - Other assets are lent and Level 1 assets are borrowed: 1.00 1.00 1.00 1.00 - Other assets are lent and Level 2A assets are borrowed: 0.85 0.85 0.85 0.85 - Other assets are lent and Level 2B RMBS assets are borrowed: 0.75 0.75 0.75 0.75 - Other assets are lent and Level 2B non-RMBS assets are borrowed: 0.50 0.50 0.50 0.50 - Other assets are lent and other assets are borrowed: 0.00 0.00 0.00 0.00 *Sources: OSFI, AMF, and IMF staff calculations.* ### Appendix VI. Cash Flow Analysis Scenarios ### Appendix VI. Cash Flow Analysis Scenarios ### Run-off Rates (Deposits) - Tables present run-off rate schedules for Retail and Small Business (RSB) deposits across Weeks 1-4 and Month 2-12 for Mild and Severe Scenarios. - Selected RSB demand / notice deposit run-off rates (Weeks 1-4 / Month 2-12): - RSB Type 1 insured, stable demand deposits: 0.25% / 0.75% / 1.00% / 2.00% - RSB Type 2 insured, stable demand deposits: 0.75% / 0.75% / 1.75% / 2.50% - RSB Insured, less stable demand deposits: 1.00% / 1.50% / 2.00% / 3.50% - RSB managed by an unaffiliated third-party, demand deposits: 3.50% / 5.00% / 8.00% / 12.00% - RSB Uninsured demand deposits: 1.00% / 1.50% / 2.00% / 3.75% - RSB Rate sensitive deposits directly managed by the client - established relationship or deposit in a transactional account: 1.00% / 1.50% / 2.00% / 3.75% - RSB Rate sensitive deposits directly managed by the client - no established relationship and not in a transactional account: 2.00% / 1.50% / 4.00% / 4.00% - Selected RSB cashable term deposit run-off rates: - RSB Type 1 insured, stable, cashable term deposit: 0.25% / 0.75% / 1.00% / 2.00% - RSB Type 1 insured, less stable, cashable term deposit: 1.00% / 1.50% / 2.00% / 3.50% - RSB managed by an unaffiliated third-party, cashable term deposit: 1.75% / 3.50% / 5.50% / 8.50% - RSB Uninsured cashable term deposit: 1.00% / 1.75% / 2.00% / 3.75% - Selected RSB fixed term deposit run-off rates (Type 1 insured, stable; 30-day to >1 year): 0.25% / 0.50% / 0.75% / 1.50% - RSB fixed term deposits managed by an Unaffiliated third-party sourced (30-day to >1 year): 1.75% / 3.50% / 5.50% / 8.50% - Commercial, Corporate and Wholesale deposit run-off highlights: - CommCorp and Wholesale Demand/Notice Deposits (Original Term ≤30 Days) - Operational: - Insured, within approved jurisdiction: 0.75% / 3.00% / 3.50% / 7.00% - Insured, outside of approved jurisdiction: 1.25% / 3.25% / 3.75% / 7.50% - Not Insured: 2.50% / 3.25% / 4.00% / 7.50% - CommCorp and Wholesale Demand/Notice Deposits (Original Term ≤30 Days) - Non-Operational: - Insured (FI): 7.50% / 25.00% - Uninsured (FI): 12.50% / 25.00% - CommCorp and Wholesale Notice, where withdrawal notification has been provided: 100.00% / 100.00% / 100.00% / 100.00% - CommCorp Notice (original term >30 days) Operational & Non-Operational: 20.00% / 20.00% / 50.00% / 50.00% - Wholesale Notice (Sovereigns, central banks, PSE, MDB): 20.00% / 20.00% / 50.00% / 50.00% - Wholesale Notice (All other counterparties, including other FIs and other legal entities): 100.00% / 100.00% / 100.00% / 100.00% - CommCorp Non-Operational Term: 20.00% / 20.00% / 50.00% / 50.00% - Wholesale Term (All other counterparties): 100.00% / 100.00% / 100.00% / 100.00% *Source: IMF staff calculations.* --- ### Haircuts (Table 2) — Securities - Haircuts specified for Mild and Severe scenarios across broad asset classes. - Government Securities (selected entries): - Sovereign & Central Bank Government Securities (High rated): 0.5% / 20.0% - State, Provincial & Agency Government Securities (High rated): 1.5% / 40.0% - State Municipal Government Securities (High rated): 5.0% / 60.0% - Supranational and Multilateral Development Bank Government Securities (High rated): 5.0% / 32.0% - Sovereign & Central Bank Government Securities (Medium rated): 10.0% / 30.0% - State, Provincial & Agency Government Securities (Medium rated): 13.0% / 100.0% - Sovereign & Central Bank Government Securities (Low/not rated): 40.0% / 100.0% - Mortgage Backed Securities (MBS): - Agency MBS (High rated): 4.0% / 15.0% - Agency MBS (Medium rated): 100.0% / 100.0% - Non-Agency CMBS (High rated): 20.0% / 100.0% - Non-Agency RMBS (High rated): 20.0% / 100.0% - Corporate Bonds and Paper (selected entries): - Non-FI issued unsecured bonds and paper (High rated): 5.0% / 60.0% - Non-FI issued covered bonds (High rated): 5.0% / 40.0% - FI issued unsecured bonds and paper (High rated): 9.0% / 80.0% - FI issued covered bonds (High rated): 9.0% / 70.0% - FI issued jumbo covered bonds (High rated): 10.0% / 100.0% - Non-FI issued unsecured bonds and paper (Medium rated): 10.0% / 70.0% - FI issued unsecured bonds and paper (Medium rated): 11.0% / 100.0% - Non-FI issued unsecured bonds and paper (Low/not rated): 60.0% / 100.0% - FI issued unsecured bonds and paper (Low/not rated): 100.0% / 100.0% - Asset Backed Securities (ABS) and Asset Backed Commercial Paper (ABCP): - Non-FI Issued ABS (High rated): 10.0% / 100.0% - Non-FI Issued ABCP (High rated): 7.5% / 40.0% - FI Issued ABS (High rated): 2.0% / 100.0% - FI Issued ABCP (High rated): 7.5% / 80.0% - Non-FI Issued ABS (Medium rated): 40.0% / 100.0% - FI Issued ABS (Medium rated): 60.0% / 100.0% - Non-FI Issued ABS (Low/not rated): 70.0% / 100.0% - FI Issued ABS (Low/not rated): 100.0% / 100.0% - Bank's own securities not eliminated: - Bank's own debt not eliminated: 100.0% / 100.0% - Bank's own equity not eliminated: 100.0% / 100.0% *Source: OSFI and IMF staff calculations.* --- ### Draws on Commitments (Table 3) — Off-Balance Sheet Items - Commitments and undrawn facility draw rates under Mild and Severe scenarios. - Retail and small business undrawn amounts (selected entries): - HELOCs: 2.00% / 10.00% - Credit Cards; of which to transactors: 0.00% / 4.00% - Credit Cards; of which to non-transactors: 1.00% / 5.00% - Other lines of credit; of which transactors: 0.00% / 4.00% - Other lines of credit; of which uncommitted to non-transactors: 1.00% / 5.00% - Other lines of credit; of which committed to non-transactors: 2.50% / 7.50% - Other; of which uncommitted: 1.00% / 5.00% - Other; of which committed: 2.50% / 7.50% - Committed credit facilities to other customers (selected entries): - Committed credit facilities to non-financial corporates; NIG; no operational relationship: 7.50% / 22.50% - Committed credit facilities to non-financial corporates; IG; no operational relationship: 7.50% / 22.50% - Committed credit facilities to non-financial corporates; NIG; existing operational relationship: 2.50% / 7.50% - Committed credit facilities to sovereigns, central banks, PSEs and MDBs: 5.00% / 15.00% - Committed credit facilities to banks subject to prudential supervision: 20.00% / 60.00% - Committed credit facilities to other financial institutions: 20.00% / 60.00% - Committed credit facilities to other legal entities: 100.00% / 100.00% - Committed credit facilities to commercial clients: - No operational relationship: 5.00% / 15.00% - Existing operational relationship: 2.50% / 7.50% - Uncommitted credit facilities to other customers: 2.50% / 7.50% (applies to non-financial corporates; sovereigns, central banks, PSEs and MDBs; banks subject to prudential supervision; other financial institutions; other legal entities) - Liquidity facilities to ABCPs: - Committed liquidity facilities to ABCPs - Maturing Balance: 100.00% / 100.00% - Committed liquidity facilities to Unissued ABCPs (reported recognizing notice periods): 100.00% / 100.00% - Undrawn amounts related to liquidity facilities to other customers: - Committed liquidity facilities to non-financial corporate clients: 15.00% / 45.00% - Committed liquidity facilities to sovereigns, central banks, PSEs and MDBs: 15.00% / 45.00% - Committed liquidity facilities to banks subject to prudential supervision: 20.00% / 60.00% - Committed liquidity facilities to other financial institutions: 100.00% / 100.00% - Committed liquidity facilities to VIEs: 100.00% / 100.00% - Committed liquidity facilities to other legal entities: 100.00% / 100.00% - Uncommitted liquidity facilities: 4.00% / 15.00% - Contingent funding obligations: - Trade finance-related obligations (including guarantees and letters of credit): 1.50% / 4.50% - Guarantees and letters of credit unrelated to trade finance obligations: 2.50% / 7.50% - Funding guarantees to subsidiaries: 100.00% / 100.00% *Source: OSFI and IMF staff calculations.* --- ### Appendix VII. Household Analysis Methodology ### A. Data Processing and Filtering - Data source and sample: - Dataset: September 2023 version of the RESL database. - Initial coverage: 9.1 million loan observations. - Final mortgage sample after filtering: 2.8 million mortgage loan observations. - Employment status coverage and exclusions: - Primary sample: borrowers employed on a full-time salaried basis comprise 48 percent of the dataset. - Expanded included employment types (part-time, hourly, seasonal, contract-based, self-employed, and unemployed at origination) bring total sample coverage to 59.1 percent. - Excluded categories: homemakers (0.2 percent), investors (0. 2 percent), and students (0.1 percent). - "Other" category excluded due to heterogeneity: 34.5 percent. - Retirees and pensioners represent 6.1 percent (initially excluded; later used in robustness checks). - Property tenure filtering: - Sample restricted to freehold and condominium/strata properties, together accounting for 95.3 percent of the dataset. - Life lease properties excluded: 0.18 percent. - Loans associated with properties categorized as "Other" excluded: 4.5 percent. - Cooperative housing observations negligible and not included. - Data cleaning and plausibility constraints: - Removed observations with missing or negative interest rates, qualifying income, property values, days past due, or current authorized amounts. - Removed loans with combined LTV ratios outside the range of (0,2). - Constrained TDS and GDS ratios at origination to the range of (0,1). ### B. Adjustment and Bring Forward of Loan at-Origination Variables - Standardization steps: - Loan payments annualized by reported payment frequency (weekly, biweekly, accelerated biweekly, monthly, semi-monthly). - Borrower income adjusted to end-2023 using employment status and loan origination date; employment status assumed unchanged from origination to end-2023. - Stochastic adjustments introduced to bring forward loan-level information to 2024 calibrated to realized macro variables. - Income projection rules: - Employed borrowers: income projected forward using wage growth trends. - Unemployed borrowers: income adjusted forward using CPI growth to approximate changes in unemployment benefits. - Income at time t adjusted using g_F(τ) where g_F(τ) refers to wage growth for employed and CPI growth for unemployed, with p representing loan-level observation. ### C. Simulation Process - Projection horizon and scenarios: - Projections over the period 2025–27, with 2024 serving as the base year. - Macroeconomic drivers: wage growth, unemployment rate dynamics, interest rate changes, housing price fluctuations. - Simulation framework: - Stochastic, loan-level simulation framework differentiates between fixed-rate and variable-rate mortgages, and between variable-rate loans with variable payments and those with fixed payments. - Bootstrap steps summary: 1) Initialize borrower status including employment status, income and DSR at t=0 where RESL data ends. 2) Iterate for each simulation for employment status, introducing idiosyncratic uncertainty via uniform random draw ξ_i ~ U(0,1). Employment transitions assigned probabilistically to match projected unemployment rate U(t). 3) Update income Y_i(t) conditional on employment status: - Employed: income grows according to projected wage growth g_w(t). - Unemployed: income declines to a fraction of prior income approximated by the OECD-reported Canadian replacement rate λ. - Transition rules: i. Employed at t−1 becoming unemployed at t: income adjusted downward using λ. ii. Unemployed at t−1 gaining employment at t: income restored by λ, assuming re-employment at prior earnings levels. iii. If employment state unchanged: income evolves based on macro wage growth. - Recalculate mortgage payments P_i(t) using loan renewal timing and projected interest rates. 4) Compute updated DSR and LTV ratio for each loan each year: - DSR computed as ratio of annualized debt payments to annual income, incorporating stochastic and scenario-specific parameters. - LTV computed as outstanding loan balance divided by estimated property value under the scenario. 5) Run 1,000 stochastic simulations per period per scenario. 6) Repeat for each year in the projection horizon under baseline and adverse macroeconomic scenarios. 7) Compute and aggregate expected DSR distribution across all simulations and loans. - Notes and limitations: - Alternative micro-simulation specifications were explored (debt service, income, employment status, LTV, debt-to-income ratios) but proved ineffective due to absence of financial savings stock data, a key determinant of default risk following unemployment. These constraints were discussed with authorities and results interpreted with caveats. ### D. Bayesian Model Averaging Step for Probability of Default Projection - The methodology indicates a Bayesian Model Averaging step is used for Probability of Default projection (section heading present; detailed specification not included in supplied content). - Replacement rate calibration: - Calibrated using unemployment benefit shares of previous income as of 2023, derived from OECD statistics calculated as the average of the replacement rate after two months (64 percent) and after one year (22 percent). *Source: Appendix VI and Appendix VII, IMF staff calculations (OSFI and RESL data as indicated in source tables).* ### 9. A by-bank panel version of the BMA framework was employed to enhance robustness ### 1canea2025008-source-pdf - 9. A by-bank panel version of the BMA framework was employed to enhance robustness ### Mortgage BMA panel specification and estimation design - A by-bank panel version of the BMA framework was employed to enhance robustness in variable selection and account for cross-sectional heterogeneity. - The estimation excludes the pandemic period (2020: Q1–2021: Q2) to mitigate distortions from temporary policy interventions and economic volatility. - The model is anchored to the 2024: Q4 DSR from Statistics Canada, incorporating projected DSR changes for 2025–27 from the structural approach as key inputs for the econometric analysis. - Model space restrictions and methodology: - Maximum of three right-hand-side variables per equation to ensure parsimony and interpretability. - Predictor pool comprises 24 variables, resulting in an implied model space of 12,950 equations. - The BMA methodology follows Gross and Población (2019, JFSR), consistent with FSAPs and euro area-wide stress testing at the ECB. - Data sources cited: BoC, Haver, Statistics Canada, OSFI, RESL, and IMF staff calculations. ### Corporate analysis methodology — data construction and sample - Firm-level dataset from Moody’s Orbis (updated through 2022) with cleaning procedures following Kalemli-Ozcan et al. (2015). - Filters applied: - Exclusion of financial sector firms. - Positive values required for total assets, cash, liabilities, debt, sales, and employment. - Only firms with valid financial statements retained. - Final sample and coverage: - Final dataset includes 1,262 firms as of 2022. - Total assets of $2.932 trillion USD in the sample. - Coverage of approximately 41 percent of total NFC assets in Canada ($7.196 trillion USD, Statistics Canada). - Public firms account for 93 percent of the sample each year on average. ### Descriptive statistics and industry structure - Industry shares by firm count: - Manufacturing: 30.6 percent. - Other industries: 20.9 percent. - Mining: 18.6 percent. - Industry shares by total assets: - Real estate: 19.7 percent. - Manufacturing: 19.3 percent. - Utilities: 12.4 percent. - Mining: 10.7 percent. - Transportation: 9.1 percent. - Information and communication: 7.3 percent. - Structural shifts (2008–2022): - Real estate sector’s share of total corporate assets increased by 17.4 percentage points. - Mining sector’s share declined by 9.7 percentage points. ### Firm-level panel econometric model specification - Dependent variable: Moody’s KMV 1-year EDF (corporate PD). - Sample restricted to publicly listed firms (Canadian SMEs underrepresented in Orbis). - Core explanatory variables aligned with stock- and flow-oriented Merton framework: - Leverage: debt-to-asset ratio where debt = 0.5*long term debt + short term debt. - Five different leverage measures (debt, asset, equity metrics) used for robustness. - ICR: EBIT over interest expense (expected negative relationship). - EBIT Ratio: EBIT over total assets (expected negative relationship). - Cash Buffer: cash holdings over short-term debt (expected negative relationship). - Cash Buffer Interactions: interaction terms with financial health indicators (e.g., ICR, EBITR, ROA) tested across specifications. - Fixed effects and estimation details: - 훼훼푖푖 represents firm-fixed effects to control for time-invariant firm characteristics. - 훾훾푡푡 represents year-fixed effects included as a robustness check. - 휀휀푖푖,푡푡 denotes the error term. - Regression estimated using both raw-level and standardized variables; standardization involves unit-specific demeaning followed by global standard deviation normalization. - Note on leverage metric terminology: This leverage metric is referred as leverage with risk weights in regression result tables. ### Comparing IMF and Bank of Canada regression measures - Broad alignment between BOC measures and IMF measures: - Cash buffers and their interaction with profitability are the most significant drivers of default risk across industries under both measures. - Sectoral heterogeneity persists: leverage is a key risk factor in capital-intensive sectors (real estate, construction, manufacturing); liquidity constraints more critical in trade and services. - Differences: - Magnitude and significance of certain coefficients differ across specifications, especially for leverage and profitability measures. - Leverage impact appears stronger in construction and manufacturing under BOC measures. - Effects of cash buffers and profitability interactions vary slightly across industries between specifications. - Overarching conclusions remain robust: sector-specific nature of financial vulnerabilities and critical role of liquidity buffers. ### Corporate Probability of Default projection using Bayesian Model Averaging - Two-step approach: 1. Firm fixed-effects panel econometric model to identify PD drivers. 2. BMA framework to project corporate PD at aggregate and sector levels using macro variables and historical NFC PD path provided by authorities. - BMA implementation details: - Methodology follows Gross and Población (2019, JFSR). - Left-hand side: aggregate and industry-level PD estimates provided by national authorities. - Right-hand side: set of 23 macroeconomic indicators. - Maximum of five variables included in each estimated model. - Pandemic period excluded from estimation (2020: Q1–2021: Q2). - Explores 44,551 possible model specifications. - Industry-level PD projections generated to assess sectoral vulnerabilities. - Data sources cited for BMA specifications: BOC, Haver, Statistics Canada, and IMF staff calculations. *Sources: Moody’s Orbis and IMF staff calculations.* --- _Source: https://www.imf.org/-/media/files/publications/cr/2025/english/1canea2025008-source-pdf.pdf_