## _cr16317 - EXECUTIVE SUMMARY ## Source details **Canonical URL:** [_cr16317 - EXECUTIVE SUMMARY](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16317.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16317.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16317.pdf.json) --- ### Inter-Sectoral Interconnectedness Analysis - Cross-border interlinkages via Irish-domiciled funds industry and multinational companies are a key feature of the financial network. - Ireland is one of a number of significant hubs for the global funds industry; tight linkages exist between the Rest of the World (ROW) and non-financial corporations reflecting large presence of foreign-controlled multinational companies. - Households: direct bilateral connection between the ROW and Irish households is insignificant; households are indirectly exposed to global shocks through insurance companies and pension funds. - Banking sector: - Tightly connected with the real sectors of the Irish economy. - Systemically important as both sources and destinations of funding in Ireland. - Strong bilateral connections with households and non-financial corporates (NFCs). - Tight bank-sovereign linkages via banks’ government bond holdings and the government’s holdings of bank equity. - Very limited linkages with the funds industry, insurance companies, or pension funds. - Other financial intermediaries (OFIs): - Do not have direct strong financial linkages with the domestic real sectors; connections mainly reflect intra-company transactions between treasury companies and parent multinationals. - Do not directly provide loans to households to any significant degree. - Largely connected through purchases of mortgage loan securitizations and loan sales from banks. - Data and coverage: - Analysis uses Quarterly Financial Accounts (QFA) covering nine sectors: MFIs; NMIFs; OFIs; ICs; PFs; GOV; HHs; NFCs; ROW. - MFIs divided into three subsectors: domestic-controlled MFIs, foreign-controlled MFIs, and money market funds. - QFAs are not consolidated and include financial linkages within and across sectors. - Key sectoral balance-sheet statistics (as of 2015Q4): - Total financial assets: 2820 percent of GDP (€6053 billion). - Financial sectors captured: 2,033 percent of GDP, comprising: - Monetary financial institutions: 533 percent of GDP. - Non-money market investment funds: 811 percent of GDP. - Other financial intermediaries: 515 percent of GDP. - Insurance companies and pension funds: 174 percent of GDP. - Net financial asset positions: - Government: -71 percent of GDP. - Non-financial corporations: -119 percent of GDP. - Households: 92 percent of GDP. - Trends: - Outstanding balance dropped by 5 percent during 2011–13 from the 2010 peak of 2,738 percent of GDP, then exceeded that peak since 2014. - Non-money market investment fund assets grew from about 270 percent of GDP in 2009 to 811 percent of GDP at 2015Q4. - Monetary financial institutions’ assets decreased by 45 percent (from 963 percent of GDP to 533 percent of GDP). - Other financial intermediaries’ assets fell by 17 percent between 2009 and 2015Q4 (from 577 percent of GDP to 515 percent of GDP) but still account for 18 percent of total financial assets. - Insurance companies and pension funds represent 6 percent of total financial assets and liabilities and have increased somewhat since the crisis. ### Banking and Financial Network Structure (inter-sectoral linkages) - Cross-border concentration and ROW linkages: - ROW’s financial assets against Irish-domiciled entities: 2,214 percent of GDP as of 2015Q4. - MMF and Non-MMF shares issued by Irish-domiciled funds and held by non-residents: €1,748 billion or 815 percent of GDP (37 percent of ROW financial assets). - ROW holdings of debt: €1,106 billion (515 percent of GDP). - ROW holdings of equity: €1,016 billion (473 percent of GDP). - ROW financial liabilities: 2,103 percent of GDP (equities and fixed-income instruments each ≈29 percent of total liabilities; loans 16 percent). - Inter-sectoral exposures (as of 2015Q2): - Non-residents held more than 90 percent of total liabilities of MMFs and non-MMFs in gross terms. - NFCs’ liabilities to domestic sectors: 153 percent of GDP (2015Q2). - Inter-bank linkages amounted to 41 percent of GDP. - Bilateral exposures between the banking sector and OFIs: 34 percent and 14 percent of GDP (two bilateral entries). - Network centrality and propagation: - Domestic network centrality highlights ROW, NFCs, OFIs, and domestic banks. - Excluding ROW, main domestic sectors by asset size (June 2015): banking 169 percent of GDP; OFIs 117 percent of GDP; NFCs 107 percent of GDP. - Using Markov chain steady-state probabilities (2015Q2): - Relative claim matrix: NFCs receive 29 cents of every euro; domestic banks receive 18 cents; OFIs receive 18 cents. - Relative obligation matrix: domestic banks supply 24 cents of every euro; OFIs supply 16 cents; households supply 15 cents. - Key transmission channels: - Domestic banks and foreign banks’ subsidiaries hold assets and liabilities vis-à-vis ROW and are exposed to credit and funding shocks from non-residents. - Insurance companies and pension funds are net creditors to the ROW; negative external shocks could transmit via their balance sheets. - ROW holds a large amount of Irish government bonds; sharp withdrawal by non-residents could generate an interest rate shock. - Multinational companies link the Irish economy with ROW through external financing channels (inter-company loans). ### Real Estate Market Analysis — Recent Developments and Valuation - Recent developments: - RRE and CRE prices have risen rapidly in recent years; rebound more vigorous than in other countries. - RRE and CRE prices in 2015Q4 are about 35 percent and 60 percent higher than the trough in 2013Q1, respectively; both remain below peak levels by 33 percent (RRE) and 48 percent (CRE). - Housing construction: number of houses completed in 2015: 12,666 units (below estimated 25,000 required). - Non-mortgage buyers significant: about half of transactions were in cash (CBI, 2015b). - Dublin: earlier and faster recovery; annual house price growth in Dublin reached 25 percent in August 2014; Dublin growth moderated to 2.6 percent at end-2015. - House price growth slowed to 6.6 percent (y-o-y) at end-2015, down from 16 percent at end-2014. - RRE valuation: - Non-parametric metrics (price-to-income and price-to-rent) suggest RRE is close to or moderately below equilibrium as of 2015Q2. - HP filters (lambda=400,000) and band-pass filters suggest house prices are somewhat undervalued by about 9–10 percent (one-sided or two-sided HP; band-pass 8–30 year interval). - VECM (1990Q1–2015Q2) indicates a negative house price gap opened after the crisis but reduced to around -5 percent in 2015. - OLS with price-to-income pseudo error correction shows no over- or undervaluation as of 2015Q2 (R-squared 36 percent). - MRSM: probability of high growth regime increased in 2013–14 but dropped at beginning of 2015; high regime expected duration estimated four years (16 quarters). - CRE valuation: - Non-parametric: price-to-rent ratio suggests CRE was moderately overvalued as of 2015Q3; breached historical average from 2014. - Statistical filters: band-pass (8–20 year) as of 2015Q3 indicates CRE prices close to equilibrium in range +2 percent; frequency filter (8–25 year) indicates 8 percent undervaluation; HP filters show close to trend. - Econometric error-correction models (1990Q1–2015Q3): long-run R-squared 74 percent; short-run R-squared 30 percent. Models indicate CRE prices marginally undervalued in range 5–8 percent (2015Q3). - MRSMs suggest high regime probability from second half of 2013 — early warning of another prolonged boom; two-regime model estimates high regime expected duration over nine years (33 quarters). - Mixed signals across methods: error-correction → marginal undervaluation (5–8 percent); price-to-rent and MRSMs → early warning of incipient new boom. - CRE market drivers and investor behavior: - CRE total returns outperformed other countries since 2014; spread between CRE total returns and long-term sovereign bond yields reached approximately 8.5 percentage points. - Low 10-year sovereign yields and supply constraints (CRE stock under construction fell dramatically 2008–2014) contributed to strong returns. - Investor base shifted to foreign investors and equity funds (REITs); foreign share of CRE transaction value rose from 2 percent in 2006 to over 40 percent in 2014 (CBI, 2015a). - New lending to CRE has slowly picked up while total outstanding CRE loan balance decreased due to deleveraging. - Policy implications (real estate): - Early warning signals of new CRE imbalances should not be overlooked; a collapse in CRE values could hit lenders with remaining exposures via financial “decelerator” mechanisms. - Authorities should continue close monitoring of CRE lending and evaluate early signals; enhance data collection and be ready to use policy toolkit. - Central Bank of Ireland and NAMA announced co-funding of a CRE statistical system by 2018 to be maintained by CSO. ### Household Sector — Deleveraging, Vulnerabilities, and Loan-Level Findings - Balance sheet and indebtedness (2008Q3–2015Q4): - Household financial assets increased by 20 percent from €294 billion to €355 billion. - Household debt fell by 27 percent from €204 billion to €150 billion. - Household net financial assets: €197 billion (92 percent of GDP) as of 2015Q4. - Household debt outstanding at €150 billion in 2015Q4 (155 percent of household disposable income and 70 percent of nominal GDP). - Household sector became a net lender in 2009 and has remained so. - Debt as proportion of total assets decreased from 28 percent in 2012Q1 to 19 percent in 2015Q4. - Composition: - Mortgage loans account for almost 90 percent of total household loans. - As of 2015Q4, deposits and claims on pension funds accounted for about 70 percent of household financial assets; remainder comprised equity holdings and claims on life insurance companies. - Mortgages equal 47 percent of GDP; mortgages equal 62 percent of household liabilities; mortgages amounted to 31 percent of bank assets. - Interest-rate exposure (2015Q3 / 2015Q4): - Rate types market share (2015Q3): - PDH: Standard Variable Rate 34.1; Tracker 36.7; Fixed less than 1 year 1.9; Fixed over 1 year 7.3; Sum 80.0. - BTL: Standard Variable Rate 6.0; Tracker 13.7; Fixed less than 1 year 0.0; Fixed over 1 year 0.2; Sum 20.0. - Interest rates (2015Q4): - PDH: Standard Variable Rate 3.96; Tracker 1.07; Fixed 1 to 3 years 3.83; Fixed Over 3 year 3.79. - BTL: Standard Variable Rate 4.61; Tracker 1.1; Fixed 1 to 3 years 4.68; Fixed Over 3 year 4.29. - About 91 percent of mortgage loans have variable interest rates. - Tracker loans and SVR loans accounted for 50.4 percent and 40.1 percent of total mortgage loans, respectively; share of loans with fixed rates over 1 year was 7.5 percent. - Average interest rate on SVR loans at 2015Q4: 3.96 percent (PDH). Interest rates on existing tracker mortgages are about 1 percent. - Negative equity and arrears (December 2015 / 2015Q4): - Fifteen percent of PDH mortgages remained in negative equity; 23 percent of BTL mortgages remained in negative equity. - PDH: 10 percent in negative equity and performing; 5 percent in negative equity and non-performing. - BTL: 12 percent in negative equity and performing; 11 percent in negative equity and non-performing. - NPL ratio declined from post-crisis peak 25.6 percent in 2014Q1 to 20.2 percent in 2015Q4. - Number and volume of mortgage arrears fell by about 36 percent and 35 percent since 2013Q3, respectively; but share of arrears >720 days-past-due rose to about 85 percent of total arrears as of 2015Q4. - Distributional points: - 43 percent of Irish households do not hold any debt. - Indebtedness is particularly heavy for the 34–45 age group. - Loan-level LLF sample and stylized facts (December 2014): - Sample: 515,210 mortgage loans totaling €78 billion (68.4 percent of total mortgage loans at end-2014). - Geography: Non-Dublin 72.3 percent; Dublin 27.7 percent. - Loan purpose: PDH 84.6 percent; BTL 15.4 percent. - Buyer status: Previous Owner 60.4 percent; FTB 39.6 percent. - Age distribution: Under 35 9.6 percent; 35–45 37.7 percent; 45–55 31.5 percent; 55–65 16.3 percent; 65–75 4.0 percent; 75+ 1.0 percent. - Current LTV distribution: Under 60 45.0 percent; 60–70 8.3 percent; 70–80 8.5 percent; 80–90 8.7 percent; 90–100 6.7 percent; 100–120 10.9 percent; 120–150 9.4 percent; 150+ 2.5 percent. - Interest rate type: Fixed 7.3 percent; SVR 51.1 percent; Tracker 41.7 percent. - Backward-looking default stocks (percent, end-2014; All Loans / PDH Only where given): - Total: All Loans 14.6; PDH Only 12.4. - Non-Dublin: All Loans 15.2; PDH Only 13.0. - Dublin: All Loans 13.1; PDH Only 11.0. - BTL: 26.4. - By age (All Loans / PDH Only): Under 35 10.0 / 9.9; 35–45 12.1 / 11.5; 45–55 14.0 / 12.9; 55–65 14.5 / 13.6; 65–75 17.2 / 16.6; 75+ 10.6 / 8.7. - By current LTV (All Loans / PDH Only): Under 60 7.5 / 7.0; 60–70 10.3 / 9.3; 70–80 11.5 / 10.1; 80–90 12.9 / 10.8; 90–100 16.9 / 14.6; 100–120 22.8 / 19.6; 120–150 31.2 / 26.7; 150+ 68.1 / 70.0. - By interest rate type (All Loans / PDH Only): Fixed 5.1 / 4.6; SVR 15.8 / 14.0; Tracker 14.7 / 11.9. - Forward-looking PDs under FSAP scenarios (three-year horizon, performing loans at December 2014): - Scenario macro inputs (selected, exact values preserved): - Real GDP growth: 2015 (est) 5.8; 2016 4.7; 2017 4.0; Baseline 2018 3.0; Adverse 2016 1.1; Adverse 2017 0.3; Adverse 2018 0.1. - Unemployment rate: 2015 (est) 9.5; 2016 8.5; 2017 7.8; Baseline 2018 7.2; Adverse 2016 9.6; Adverse 2017 10.5; Adverse 2018 11.2. - Short-term interest rate (3 month Euribor): 2015 (est) 0.0; 2016 0.0; 2017 0.1; Baseline 2018 0.4; Adverse 2016 0.8; Adverse 2017 0.9; Adverse 2018 1.2. - Long-term interest rate (10 year Irish government bond): 2015 (est) 1.3; 2016 1.7; 2017 2.3; Baseline 2018 3.3; Adverse 2016 3.2; Adverse 2017 5.3; Adverse 2018 7.9. - Residential property prices: 2015 (est) 6.5; 2016 7.5; 2017 6.7; Baseline 2018 5.1; Adverse 2016 -5.0; Adverse 2017 -5.0; Adverse 2018 0.0. - Commercial property prices: 2015 (est) 21.8; 2016 10.3; 2017 4.2; Baseline 2018 1.9; Adverse 2016 -4.6; Adverse 2017 -16.1; Adverse 2018 -11.2. - Key forward-looking PD findings (average three-year PD, 2014–17 under adverse FSAP): - LTV and stressed PDs: strictly increasing relationship; LTV over 150 percent shows adverse PDs climbing above 20 percent. - Dublin vs Non-Dublin: Baseline close to 6 percent; Adverse close to 8 percent. - PDH vs BTL: smaller difference in forward-looking PDs than backward-looking default stocks. - FTB vs Previous Owner: FTB adverse three-year PD over 9 percent; Previous owners close to 7 percent (driven by CLTV distributions: average CLTV 53 for previous owners vs. 75 for FTBs). - Age groups: Under 35 adverse PD 10 percent on average; 65–75 adverse PD 6 percent on average. - Interest rate type: Tracker loans projected PD 6 percent under adverse scenario; SVR loans projected PD closer to 10 percent under adverse scenario; fixed-rate loans have lower forward-looking PDs. - Distinction: largest source of mortgage credit risk in most household segments is existing stock of defaulted loans rather than probabilistic default risk among performing loans. - Policy recommendations (households and mortgage market): - Continue collecting loan-level data for systemic risk assessment. - Central Bank of Ireland should move forward with establishing the Central Credit Register (CCR) to ensure accurate and verifiable household credit information. - Maintain proportionate limits on LTV and LTI ratios to build resilience; monitor risky segments (high LTV, young borrowers, FTBs, BTL investors, SVR borrowers). ### Corporate Sector Analysis — Performance, Vulnerabilities, and Stress Tests - Recent performance and headline statistics: - 2015 corporate sector real growth rate: 12 percent. - 2015 gross operating surplus increased by 15 percent in real terms. - Profitability share reached about 65 percent versus pre-crisis average of 55 percent. - NFC sector debt: 187 percent of GDP (2015Q4). - NFC debt-to-equity ratio: 45 percent in 2015Q4 (down from peak 123 percent in early 2009). - Multinationals’ consolidated debt: 94½ percent of GDP in 2014 compared to 36 percent of GDP in 2007. - Debt of Irish firms: 85 percent of GDP in 2014 from about 65 percent of GDP in 2007. - Share of loans from non-Irish entities: 55 percent of total loans in 2015Q4 compared with 34 percent at end-2012. - Share of non-euro denominated debt in total debt increased to more than half in 2015 from about one-quarter in 2006. - Non-performing loans and distress: - Corporate NPLs: about 17 percent in 2015Q3 (peak nearly 27 percent at end-2013). - Construction sector NPL ratio remained above 50 percent. - Corporate bankruptcies: quarterly average about 300 (2009–2012) versus 84 (2005–07); recent quarterly average about 225 cases (2014–15). - BSM 4½ percent; Z-probability 14 percent in 2014 (vs. 2008 BSM 24 percent; Z-probability 32 percent). - Firm-level sample and methodology: - ORBIS dataset: unbalanced panel 1995–2014; total 9,567 observations and 3,767 firms (3,040 domestic firms: 7,663 observations; 727 foreign firms: 1,904 observations). - Firm size composition (domestic): Small 4,409 observations / 1,882 firms; Medium 2,266 / 833; Large 988 / 325. - Analysis focuses on Interest Cover Ratio (ICR = EBITDA / interest expenses); ICR < 1 indicates technical default; ICR between 1 and 2 regarded as heightened risk. - Stylized facts: - Median RoE: fell to just below 3 percent in 2009 from 14 percent in 2007. - Debt-to-equity of median firm declined to 52 percent in 2013 from 65 percent in 2007; median large firm debt-to-equity increased to 100 percent in 2013 from 75 percent in 2007. - Median firm ICR declined to about 4 in 2008–09 from just above 6 in 2007; by 2013 median domestic firm ICR 6.3; median foreign firm ICR 6.6. - Share of vulnerable domestic firms (ICR < 2): 2007 24.4 percent; 2009 37.2 percent; 2013 24.3 percent. - Share of debt owned by domestic firms with ICR < 1: 2007 10.5 percent; 2009 15.3 percent; 2013 9.9 percent. - Composition shift: in 2013 small firms accounted for about two-thirds of risky debt (vs. one-third in 2007). - Econometric distress analysis (Logit): - Sample ~9,500 observations, ~706 distress events (~7½ percent). - ICR is an important predictor of distress; current ratio and firm size reduce distress probability; construction sector more likely to be in distress. - Pseudo R2 range 0.086–0.100 across specifications. - Sensitivity analysis and stress-test calibration (static, end-2013 balance sheets; assumptions preserved): - Three shocks assessed: Interest rate shock (450bp), Profit shock (15 percent decline in added value), Combined shock. - Assumptions: - Apply 450bp shock to effective interest rate; assume 50 percent of 2013 debt stock rolled-over at higher rate. - Firms’ financial assets assumed 10 percent of total assets. - Basel standard LGD of 45 percent applied in loss calculations; alternate LGD 60 percent also examined. - Assumed bank ownership of firm debt: 50 percent of large firms’ debt; 75 percent of SMEs’ debt (sensitivity uses 50 percent for SMEs). - Stress-test headline results (percent unless indicated): - Baseline new corporate defaults: 3.6 percent of overall bank corporate loan book (cumulative, two-year horizon). - Profit shock: new defaults 5.7 percent of bank corporate loan book. - Interest rate shock: new defaults 7.1 percent of total corporate loans. - Combined shock: new defaults 9.4 percent of total corporate loans. - Impact on banks’ regulatory Tier 1 capital: - Baseline (2015Q3) Tier 1: 17.4 percent of risk-weighted assets. - Combined shock (LGD = 45 percent): Tier 1 falls to 15.1 percent. - Combined shock (LGD = 60 percent): Tier 1 falls to 14.3 percent. - Re-weighted sample toward higher SME representation (combined shock): Tier 1 falls to 14.5 percent from 17.4 percent. - Interpretation: - Interest rate shock would push median ICR from 6.8 to 4.0 and increase share of firms with ICR < 2 to nearly one-third; share of risky debt (debt held by firms with ICR < 2) would rise to 56.6 percent. - Profit shock would bring median ICR to 2.0 and increase share of firms with ICR < 1 to 42.9 percent. - Combined shock would raise share of firms with ICR < 1 to 47.4 percent and share of firms with ICR < 2 to 58.6 percent; share of debt ICR < 2 to 59.9 percent. - Under these static first-round exercises banks’ Tier 1 capital would remain above minimum requirement, but second-round effects and wider macro impacts could amplify losses. ### Data Collection, Monitoring, and Policy Recommendations - Data initiatives and gaps: - CBI: large exposures data for bilateral interbank exposures; security-level granular data from banks and non-money market investment funds quarterly; MMF monthly reporting from December 2014; FVC reporting extended to other SPVs from 2015Q3; insurers’ detailed data under Solvency II from 2016. - Liability-side interlinkages to be monitored with new EU Commission data template (being finalized). - Current limitation: granular bilateral exposure data for banks mainly available for asset exposures. - Policy implications and recommendations (summarized): - Enhance collection of granular bilateral exposure data within banking and non-bank financial institutions and across sectors; allocate sufficient resources and inter-departmental collaboration for data initiatives. - Maintain and monitor macroprudential measures on mortgage lending (LTV and LTI limits); evidence suggests these measures may have moderated price pressures. - Continue intensive monitoring of CRE lending; enhance CRE data collection (CRE statistical system co-funded by CBI and NAMA, to be maintained by CSO by 2018). - Central Bank of Ireland should proceed with establishment of the Central Credit Register (CCR) and consider expanding loan-level data collection to all regulated lenders if market lending institutions increase. - Monitor risky corporate segments and prepare contingency planning given corporate sector’s external funding reliance and potential for reversals in global market sentiment. *Source: EXECUTIVE SUMMARY and chapter content, _cr16317 (Ireland Financial Sector Assessment Program), International Monetary Fund* ### EXECUTIVE SUMMARY ___________________________________________________________________________ 7 ### _cr16317 - EXECUTIVE SUMMARY ### Inter-Sectoral Interconnectedness Analysis - Cross-border interlinkages via Irish-domiciled funds industry and multinational companies are a key feature of the financial network. - Ireland is one of a number of significant hubs for the global funds industry. - Tight linkages exist between the Rest of the World (ROW) and non-financial corporations, reflecting the large presence of foreign-controlled multinational companies in Ireland. - Direct bilateral connection between the ROW and Irish households is insignificant; households are indirectly exposed to global shocks through insurance companies and pension funds. - The banking sector: - Is tightly connected with the real sectors of the Irish economy. - Is systemically important as both sources and destinations of funding in Ireland. - Has strong bilateral connections with households and non-financial corporates. - Has tight bank-sovereign linkages via banks’ government bond holdings and the government’s holdings of bank equity. - Has very limited linkages with the funds industry, insurance companies, or pension funds. - Other financial intermediaries (OFIs): - Do not have direct strong financial linkages with the domestic real sectors; connections mainly reflect intra-company transactions between treasury companies and parent multinationals. - Do not directly provide loans to households to any significant degree. - Are largely connected through purchases of mortgage loan securitizations and loan sales from banks. - Data and coverage: - This analysis uses Quarterly Financial Accounts (QFA) data covering nine sectors: (1) monetary financial institutions (MFIs) including deposit-taking institutions and money market funds; (2) non-money market investment funds (NMIFs); (3) other financial intermediaries including financial vehicle corporations and holding companies (OFIs); (4) insurance companies (ICs); (5) pension funds (PFs); (6) governments (GOV); (7) households including non-profit institutions serving households (HHs); (8) non-financial corporations (NFCs); and (9) the rest of the world (ROW). - MFIs are divided into three subsectors: domestic-controlled MFIs, foreign-controlled MFIs, and money market funds. - The QFAs are not consolidated and include financial linkages within and across sectors. - Key statistics on sectoral balance sheets (as of 2015Q4): - Total financial assets: 2820 percent of GDP (€6053 billion). - Financial sectors captured: 2,033 percent of GDP, comprising: - Monetary financial institutions: 533 percent of GDP. - Non-money market investment funds: 811 percent of GDP. - Other financial intermediaries: 515 percent of GDP. - Insurance companies and pension funds: 174 percent of GDP. - Net financial asset positions: - Government: -71 percent of GDP. - Non-financial corporations: -119 percent of GDP. - Households: 92 percent of GDP. - Trends: - Outstanding balance dropped by 5 percent during 2011–13 from the 2010 peak of 2,738 percent of GDP, then exceeded that peak since 2014. - Non-money market investment fund assets grew from about 270 percent of GDP in 2009 to 811 percent of GDP at 2015Q4 (threefold increase). - Monetary financial institutions’ assets decreased by 45 percent (from 963 percent of GDP to 533 percent of GDP). - Other financial intermediaries’ assets fell by 17 percent between 2009 and 2015Q4 (from 577 percent of GDP to 515 percent of GDP) but still account for 18 percent of total financial assets. - Insurance companies and pension funds represent 6 percent of total financial assets and liabilities and have increased somewhat since the crisis. ### Real Estate Market Analysis - Recent developments: - Residential real estate (RRE) and commercial real estate (CRE) prices in Ireland have risen rapidly in recent years, raising concerns about possible overvaluation and a new build-up of imbalances. - The rebound in Irish RRE and CRE markets has been more vigorous than in other countries. - Valuation assessments: - RRE: - Both non-parametric and parametric methods suggest the RRE market is close to or moderately below its equilibrium level. - Two standard metrics—price-to-income and price-to-rent ratios—indicate that after a period of pre-crisis overvaluation and a subsequent correction, the market has been close to its equilibrium level in recent quarters. - CRE: - Approaches send mixed signals on current CRE valuation. - The price-to-rent ratio shows that after a sharp correction at the onset of the crisis, CRE prices increased above historical averages in 2014 and were moderately overvalued. - Frequency and HP filters show CRE prices have recently converged to their long-term trends. - Foreign investment inflows or equity funding like REITs can reverse if market sentiment changes, potentially causing a sharp drop in CRE values. - Risks and policy considerations: - Early signals of new imbalances in the CRE market should not be overlooked. - A collapse in CRE values could hit lenders with remaining exposures via financial “decelerator” mechanisms observed in the post-crisis period. - Authorities should continue to closely monitor CRE lending and evaluate early signals of a build-up of new market imbalances. ### Household Sector Analysis - Deleveraging and indebtedness: - Households have deleveraged but remain highly indebted. - Some mortgage loans remain in negative equity. - Almost all mortgages have variable interest rates (tracker and standard variable rates). - Vulnerabilities and risks: - Negative equity is a well-documented cause of default in addition to income shocks. - Interest rates on existing tracker mortgages are currently very low, reflecting the zero ECB policy rate; an increase in interest rates would raise interest payments and could make debt servicing difficult for some households. - Existing borrowers with high loan-to-value (LTV) ratios or standard variable rates are more vulnerable to the FSAP adverse scenario. - The LTV distribution shows a strictly increasing relationship between LTV levels and the stressed probability of default. - Data and policy recommendations: - Continued collection of loan-level data is crucial for systemic risk assessment. - The Central Bank of Ireland should move forward with establishing the Central Credit Register (CCR) to ensure individual households’ credit information can be accurately and verifiably recorded. ### Corporate Sector Analysis - Overall resilience: - Vulnerabilities among non-financial corporations (NFCs) have moderated in recent years. - The interest cover ratio (ICR) of the median firm across all categories of firm size and sectors increased steadily in recent years following a sharp decline in 2008-09. - The share of risky debt (firms with ICR lower than one) in total debt declined to the pre-crisis level. - Remaining vulnerabilities: - The NFC sector—especially smaller firms—remains highly vulnerable. - Small domestic firms account for most firms under “technical default” (ICR less than one). - For small firms, the share of risky debt constituted nearly half of their total debt, well above shares for medium-sized and large enterprises. - Sensitivity analysis and stress scenarios: - The sector, particularly smaller firms, remains highly vulnerable even to non-extreme shocks. - Ceteris paribus, banks’ regulatory capital would still remain above the minimum requirement under the analyzed scenarios. - An adverse shock comprising a decline in profitability and an increase in interest rates would likely: - Triple the share of firms with ICR lower than one to nearly fifty percent, largely reflecting deterioration among small firms. - Increase the share of risky debt to levels observed during the financial crisis, resulting in a significant increase in new corporate defaults. - Given the moderate share of corporate loans on banks’ loan books and banks’ comfortable capital positions, banks would still be able to keep their regulatory Tier 1 capital well above the minimum requirement under the analyzed adverse scenario. *Source: EXECUTIVE SUMMARY, _cr16317 (Ireland Financial Sector Assessment Program), International Monetary Fund* ### 6. For most financial sectors, assets are mainly allocated to equities, long-term debt ### 6. For most financial sectors, assets are mainly allocated to equities, long-term debt securities, or long-term loans, but are funded by different instruments ### Asset and liability composition by sector (summary findings) - As of 2015Q3, monetary financial institutions (MFIs) held more than a half of their financial assets in liquid form (cash and deposits or short-term debt securities, e.g., certificates of deposit, commercial paper, treasury bills, etc.). - MFIs had two main sources of funding: deposits for retail banks and credit unions, and shares for money market funds (MMFs). - Non-money market investment funds (non-MMIFs) assets consisted primarily of long-term debt securities, equities, and financial derivatives; 77 percent of their liabilities were fund shares. - Other financial institutions (OFIs) asset composition was dominated by loans (mostly long-term); OFI liabilities came from long-term debt securities, equities, and loans. - Insurance companies’ liabilities were mostly insurance premia; they held non-MMF shares and long-term debt securities. - Pension funds’ assets consisted of equities and long-term debt securities; their liabilities were completely funded by pension entitlements. ### Household sector: deleveraging and asset composition - Household outstanding debt decreased by 27 percent (-€54 billion) between 2008 and 2015Q4; almost all household debt was long-term mortgage loans. - Household financial assets remained broadly stable through increases in pension entitlements offsetting decreases in deposits, equity holdings, and claims on insurance companies. - As of 2015Q4, deposits and claims on pension funds accounted for about 70 percent of household financial assets; the remainder comprised equity holdings and claims on life insurance companies. - The household sector was a net lender, with net financial assets amounting to €197 billion (92 percent of GDP). ### Non-financial corporations (NFCs): balance sheet growth and composition - Corporate financial assets increased 2.4 times in the post-crisis period; corporate liabilities doubled over the same period. - Equities had the largest share of both assets and liabilities of NFCs (51 percent and 60 percent), largely due to new entries of multinational companies. - Corporate debt increased from €323 billion to €402 billion (an increase of 15 percentage points of GDP) between 2008 and 2015Q4. - External debt, mainly owned by multinational firms, stood at €214 billion in 2015Q4, accounting for more than 50 percent of total corporate debt in Ireland. - Note: CSO is working on separating data on multinational firms’ debt from Irish domestic firms. ### Rest of the World (ROW) exposures and cross-border linkages - As of 2015Q4, the ROW’s financial assets against Irish-domiciled entities and residents represented almost 2,214 percent of GDP. - MMF and Non-MMF shares issued by Irish-domiciled funds and held by non-residents amounted to €1,748 billion or 815 percent of GDP; these accounted for 37 percent of total ROW financial assets. - ROW holdings of debt and equity were €1,106 billion and €1,016 billion (515 and 473 percent of GDP), accounting for 23 percent and 21 percent of the ROW’s financial assets, respectively. - ROW financial liabilities were 2,103 percent of GDP and were mainly composed of equities, fixed-income instruments, and loans (the first two each comprise about 29 percent of total liabilities, and loans comprise 16 percent of the total). - The volume of ROW assets and liabilities vis-à-vis Irish-domiciled entities is exceptionally large relative to other European countries (noted next to Luxembourg in the comparison). ### Inter-sectoral financial linkages and network properties (as of 2015Q2) - Cross-border interlinkages are strongest for the Irish-domiciled funds industry and Irish-domiciled multinational companies; all bilateral connections over 100 percent of GDP in 2015Q2 were between the ROW and domestic sectors. - As of 2015Q2, non-residents held more than 90 percent of total liabilities of MMFs and non-MMFs in gross terms. - The ROW has connections with all other sectors; NFCs are exposed to all counterparties except MMFs and pension funds (PFs). - Domestic financial network centrality (degree) highlights the prominence of ROW, NFCs, OFIs, and domestic banks. - Excluding the ROW, the three main domestic sectors by asset size (June 2015) were: - Banking sector (domestic banks): financial assets at 169 percent of GDP. - Other financial institutions (OFIs): 117 percent of GDP. - Non-financial corporations (NFCs): 107 percent of GDP. - Many domestic sectors had large financial claims on and obligations to these three sectors relative to their balance sheets; shocks to these central sectors can propagate through direct and indirect interlinkages. ### Key transmission channels of cross-border risks (analysis) - Domestic banks and foreign banks’ subsidiaries hold both financial assets and liabilities against the ROW and are exposed to spillovers from credit and funding shocks faced by non-residents; however, such spillovers are less concerning than in the pre-crisis period because foreign banks’ exposures to Ireland and Irish domestic banks’ claims on non-residents have fallen significantly in recent years. - Irish insurance companies and pension funds are net creditors to the ROW; negative shocks originating outside Ireland could be transmitted through their balance sheets. - The ROW holds a large amount of Irish government bonds; a sharp withdrawal of non-residents from the bond market could generate an interest rate shock and distress in the Irish financial system. - Multinational companies link the Irish economy with the ROW through external financing channels (e.g., inter-company loans). ### Selected exact statistics and magnitudes (preserved) - Household net financial assets: €197 billion (92 percent of GDP) as of 2015Q4. - Household outstanding debt change: -27 percent (-€54 billion) between 2008 and 2015Q4. - Corporate debt: rose from €323 billion to €402 billion (increase of 15 percentage points of GDP) between 2008 and 2015Q4. - External corporate debt (mainly multinationals): €214 billion in 2015Q4 (>50 percent of total corporate debt). - ROW financial assets against Irish-domiciled entities: 2,214 percent of GDP as of 2015Q4. - MMF and Non-MMF shares held by non-residents: €1,748 billion or 815 percent of GDP (37 percent of ROW financial assets). - ROW holdings of debt: €1,106 billion (515 percent of GDP). - ROW holdings of equity: €1,016 billion (473 percent of GDP). - ROW financial liabilities: 2,103 percent of GDP (equities and fixed-income instruments each ≈29 percent of total liabilities; loans 16 percent). - As of 2015Q3, MFIs: more than a half of financial assets in liquid form. - Non-MMIFs: fund shares comprised 77 percent of liabilities. - Domestic sector asset sizes (June 2015): banking 169 percent of GDP; OFIs 117 percent of GDP; NFCs 107 percent of GDP. - Cross-border concentration as of 2015Q2: non-residents held more than 90 percent of total liabilities of MMFs and non-MMFs. *Source: IMF staff compilation and calculations based on CBI and CSO data (figures and tables as of 2015Q2–2015Q4 in the source document).* ### 13. The Irish banking sector is tightly connected with the real sectors of the Irish economy, ### 13. The Irish banking sector is tightly connected with the real sectors of the Irish economy, ### Inter-sectoral financial linkages: summary findings - Banks’ obligations were well diversified, with households playing the largest role (over 30 percent of bank liabilities and household assets). - Destinations of banks’ claims were mainly the real sectors, including households and NFCs (referenced in Table 5 and Table 6 in source). - Tight bank–sovereign linkage via banks’ government bond holdings and the government’s holdings of bank equities. - Inter-bank linkages amounted to 41 percent of GDP (red box in Table 4). - Bilateral exposures between the banking sector and OFIs amounted to 34 percent and 14 percent of GDP (blue boxes in Table 4). ### Other financial intermediaries (OFIs), insurance companies, and pension funds - Irish funds industry and the banking system did not depend on each other for funding and asset allocation. - Insurance companies’ exposure to banks: 4 percent of GDP. - Pension funds’ exposure to banks: 1 percent of GDP. - Banks had no meaningful exposure to insurance companies and pension funds (Table 4). - OFIs held 36 percent of their total assets in the form of NFC equity or debt. - NFCs invested 14 percent of assets into OFIs. - OFIs had exposures to households equal to 19 percent of their total assets. - OFIs do not provide loans directly to households to any significant degree; connections are largely through purchases of mortgage loan securitizations and loan sales from banks. - The connection between OFIs and the domestic real sector mainly reflects intra-company transactions between treasury companies and their parent multinational companies. ### Non-financial corporates (NFCs) - NFCs are a net borrower and the most indebted sector in Ireland. - In 2015Q2, NFCs’ liabilities to domestic sectors were 153 percent of GDP (Table 4). - Most of the claims on NFCs were held by: - the non-financial corporate sector itself: 60 percent of GDP, - OFIs: 42 percent of GDP, - banks: 25 percent of GDP. - Role of treasury companies within multinational groups is significant (see text). ### Household sector - Households are the largest net lender in Ireland. - Households’ retirement savings in pension funds: 33 percent of household financial assets. - Households’ holdings in banks’ deposits or debentures: 32 percent of household financial assets. - Households’ claims on insurance companies: 23 percent of household financial assets. - Largest claim on households was held by banks in the form of mortgages: - Mortgages equal 47 percent of GDP, - Mortgages equal 62 percent of household liabilities, - Mortgages amounted to 31 percent of bank assets. ### Comparative perspective: Ireland vs Denmark (end-2013) - Denmark (end-2013): interbank claims were 110 percent of GDP; institutional investors–OFIs linkages were 55 percent of GDP. - Ireland (2015Q2): interbank claims amounted to 47 percent of GDP. - Bilateral exposures between institutional investors and OFIs and NMIFs in Ireland: 14 percent of GDP. - Bilateral linkages between domestic financial sectors and real sectors in Denmark (end-2013): - Banks’ loans to households: 129 percent of GDP, - Banks’ loans to NFCs: 64 percent of GDP. - Comparable Ireland figures (2015Q2): - Banks’ loans to households: 49 percent of GDP, - Banks’ loans to NFCs: 26 percent of GDP. ### Network analysis (Markov chain steady-state probabilities) - Using the relative claim matrix (Table 6) as transition matrix, steady-state destination probabilities (2015Q2) indicate: - NFCs receive 29 cents of every euro circulating through the domestic financial network. - Domestic banks receive 18 cents of every euro. - OFIs receive 18 cents of every euro (text implies OFIs among second largest funding destinations). - Using the relative obligation matrix (Table 7) as transition matrix, steady-state source probabilities (2015Q2) indicate: - Domestic banks supply 24 cents of every euro when tracing funding back. - OFIs supply 16 cents of every euro. - Households supply 15 cents of every euro. - Non-money market investment funds are important international funding sources but mainly affect the domestic economy through economic channels (employment, investment) rather than direct financial links. - Implication: income/profit or liquidity shocks to NFCs and domestic banks can affect upstream funding sources; liquidity shocks to domestic banks can significantly affect funding positions of downstream destinations. ### Data collection, monitoring, and gaps - Central Bank of Ireland (CBI) efforts to fill data gaps: - Large exposures data used to examine bilateral interbank exposures (Hallissey, 2016). - Liability-side interlinkages to be monitored with a new EU Commission data template (being finalized). - Studies of interconnectedness between non-domestic banks and Irish-domiciled counterparties in CDS market (Kenny and others, 2015) using EMIR derivative market data. - Security-level granular data collected quarterly from banks and non-money market investment funds; less granular FVC data in line with ECB regulations. - CBI extended security-level reporting to MMFs on a monthly basis from December 2014. - CBI extended FVC reporting form in full to other Special Purpose Vehicles from 2015Q3 (an internationally unique measure). - New detailed exposure data of insurance companies will be available in 2016 under Solvency II reporting regime. - Current limitation: granular bilateral exposure data for banks are currently available mainly for asset exposures. ### Conclusions and policy implications (key points) - The Irish economy is highly connected across its border; despite deleveraging, banks remain interconnected with the rest of the world (ROW). - If banks lack sufficient capital and liquidity buffers, they can act as shock amplifiers. - Insurance companies and pension funds can propagate negative foreign shocks into Ireland because domestic households are exposed to non-residents through savings invested in these sectors. - Government and NFCs rely on funding from non-residents and remain vulnerable to reversals in global market sentiment. - Banking sector, NFCs, and OFIs (largely connected to MNC activities) are important funding sources and destinations within the domestic-domiciled network; income/profit or liquidity shocks in these sectors can generate negative spillovers unless capital/liquidity buffers are adequate. - Importance of enhancing collection of granular bilateral exposure data within both banking and non-bank financial institutions and across sectors. - Existing securities data collected for banks, investment funds, MMFs, and FVCs; SPVs added from Q3 2015; insurers’ detailed data under Solvency II from Q1 2016. - Data initiatives underway at the CBI are welcome but will require sufficient resources and strong inter-departmental collaboration. ### Real estate market analysis — overview and risks - Ireland experienced a historic financial crisis from the bursting of a real estate boom during the global financial crisis. - During 2000–2007: RRE and CRE prices doubled and total banking assets tripled. - Consequences of the boom and bust: - GDP growth was negative for two years in 2008–09. - Unemployment increased to 15.1 percent by 2012. - House prices fell by about 50 percent from their 2007 peak; CRE prices fell by about 70 percent from their 2007 peak. - Authorities recapitalized the banking system by €64 billion (about 40 percent of GDP). - System continues to carry a large stock of NPLs, significantly composed of long-overdue mortgages and CRE loans. - National Asset Management Agency (NAMA) is included in OFI sector; NAMA assets amounted to €11 billion mainly in the form of loans and other receivables, as of 2015Q3. - Post-crisis recovery in real estate markets: - RRE and CRE prices in 2015Q4 are about 35 percent and 60 percent higher than the trough in 2013Q1, respectively. - Both remain below peak levels by 33 percent (RRE) and 48 percent (CRE), respectively. - Key drivers and indicators: - Housing construction subdued; number of houses completed in 2015: 12,666 units (well below estimated 25,000 units required). - Correlation between annual house price growth and private credit growth: - 0.67 before the crisis (1990–2007Q3), - -0.04 after the crisis (2007Q4–2015Q3). - Mortgage loan approval increased since 2012H2 but nominal amount still below pre-2000 level. - Non-mortgage buyers significant: about half of transactions were in cash (CBI, 2015b). - Regional divergence: - Dublin prices recovered earlier and about three times faster than rest of country. - Annual house price growth in Dublin reached 25 percent in August 2014. - Recent moderation: - House price growth slowed to 6.6 percent (y-o-y) at end-2015, down from 16 percent at end-2014. - Dublin growth rate fell back to 2.6 percent at end-2015. - Market expectations moderated: 2015Q3 CBI survey expecting prices to rise across horizons: - 1 quarter: 46 percent, - 1 year: 82 percent, - 3 years: 93 percent; - Down from 90, 97, and 98 percent respectively in 2014Q3. - Commercial real estate (CRE) developments: - Based on MSCI/IPD, capital values grew by 30 percent (y-o-y) at end-2014 and 19 percent (y-o-y) at 2015Q4. - Rental values increased rapidly above 14 percent (y-o-y) for the last four quarters, and are 41 percent higher than the trough in 2013. - Dublin office sector performance most robust and expanding beyond the capital. - CRE capital values increased by 25 percent (y-o-y) since 2014Q2 — fastest growth since 1999. *Source: International Monetary Fund, Ireland Financial Sector Assessment Program chapter content.* ### 36. Strong returns have attracted investors in the search for yield as interest rates remain low. ### _cr16317 - 36. Strong returns have attracted investors in the search for yield as interest rates remain low. ### Commercial real estate (CRE) market performance and drivers - The Irish CRE market has been one of the best performing asset classes in Europe since 2014. - Total returns for Irish CRE have outperformed other countries (Figure 11). - Returns on 10-year Irish sovereign debt have trended to a twenty-year low and the spread between total returns for CRE and the long-term sovereign bonds reached approximately 8.5 percentage points. - The favorable economic outlook has boosted the CRE market recovery, with broad-based improvement in domestic economic performance driving rental growth in the office sector. ### Supply constraints and construction activity - Lack of new construction activity has contributed to supply shortages. - CRE stock under construction fell dramatically between 2008 and 2014, reflecting: - low profitability margins in an environment of depressed prices (early in the period); - construction firms’ stretched balance sheets and tighter lending standards by banks; - possibly other factors, such as the limited availability of suitable sites. - Activities are slowly picking up, but the historical fall in construction contributed to current shortages. ### Investor composition and motivations - The investor base has changed considerably since the crisis: majority of investment activity is now funded through foreign investors and equity funds, such as real estate investment trusts (REITs). - Large foreign pension and insurance funds from the U.S., the U.K., and Germany have invested in the Irish CRE market, in part for balance sheet management and matching long-term liabilities with long-term assets. - The attractiveness of Ireland as a leading destination for FDI has increased demand for CRE properties from multinational companies. - CBI (2015a) notes that while only 2 percent of the value of CRE transactions was from foreign investors in 2006, this figure was over 40 percent in 2014. - Property funds such as REITs allow smaller investors access to the market and additional liquidity, with listed funds offering equity securities that can be bought and sold during market hours and some unlisted funds willing to return investments at a month’s notice. ### Credit, deleveraging, and new lending - Total outstanding balance of CRE loans has been decreasing while the flow of new lending to CRE has slowly picked up. - Banks reduced exposure in the CRE market through deleveraging from 2009 to 2014. - New bank financing to the CRE sector is slowly increasing, but new lending to CRE has been a minor component of total new lending volumes. ### Empirical analysis of price misalignment in real estate markets - The report employs several non-parametric and parametric methods to estimate long-term trend or equilibrium price levels and compare with actual prices to test for overvaluation in RRE and CRE. - Non-parametric methods used: price-to-income and price-to-rent ratios. - Parametric methods used: statistical filters, time-series approaches using economic fundamentals (VECM and OLS with pseudo error correction), and Markov regime switching models (MRSM). - Estimating equilibrium levels of house prices is challenging and results should be interpreted with caution. ### Residential real estate (RRE) valuation — non-parametric findings - Common measures based on deviation from long-run historical trends (price-to-income and price-to-rent ratios) suggest the market is neither undervalued nor overvalued in 2015Q2. - Both ratios show an apparent overshooting in house prices before the crisis and a correction afterwards. - International comparison: using the price-to-rent ratio across 25 OECD countries, Ireland is close to the middle of the distribution as of 2015Q2. - Price expectations (CBI survey): percentage of respondents expecting prices to rise across horizons: - 1 quarter: 46 percent (down from 90 percent in 2014Q3) - 1 year: 82 percent (down from 97 percent in 2014Q3) - 3 years: 93 percent (down from 98 percent in 2014Q3) - House price growth rate moderated to 6.6 percent (y-o-y) at end-2015, from 16 percent at end-2014. - The Central Bank of Ireland imposed proportionate limits on loan-to-value (LTV) and loan-to-income (LTI) ratios in February 2015 with aims to strengthen resilience of lenders and borrowers and reduce risk of future credit and RRE price spirals; it is still too early to fully evaluate effectiveness, though there is some evidence these measures reduced price pressures. ### Residential real estate — parametric/statistical filter results - Statistical filters suggest house prices are somewhat undervalued by 9–10 percent: - Using either one-sided or two-sided HP filter (lambda=400,000), Irish house prices are estimated to be about 9 percent below the trend. - Isolating a component within an 8–30 year interval using a band-pass filter suggests an undervaluation of about 10 percent. ### Econometric model results (VECM, OLS, MRSM) - VECM results (quarterly data 1990Q1–2015Q2): - A negative house price gap opened after the crisis but reduced to around -5 percent in 2015. - Model uses five endogenous variables: real house prices, real GDP per capita, real private credit, long-term interest rate (10-year government bond yield), and real construction cost. - Two model specifications were estimated and averaged: one with three lags and two cointegration vectors; the other with two lags and three cointegration vectors. - OLS regression with price-to-income ratio as pseudo error correction term: - Shows no over- or undervaluation in house prices as of 2015Q2. - Model yields R-squared of 36 percent and indicates the deviation from equilibrium closed by 2015Q2. - Markov regime switching model (MRSM) findings: - MRSM allows parameters to change between regimes (high or low; sometimes three states: high, normal, low). - Probability of being in a high growth regime (boom) increased in 2013–14, but dropped at the beginning of 2015. - The high regime has occurred for half of the sample period, and its expected duration is estimated to be four years (16 quarters). - Two booms in 1998–2001 and 2002–07 stand out markedly, and one reappeared in 2014. - The MRSM with three regimes confirms the two-regime findings; probability of a high regime in both specifications started to decrease in 2015. ### Summary findings and policy-relevant observations - Non-parametric and parametric analyses suggest Irish RRE prices are currently around equilibrium levels as of 2015Q2. - Prices rebounded during 2013–14, but pace started to slow at the beginning of 2015. - Recent data show that macroprudential measures on mortgage loans (LTV and LTI limits introduced February 2015) may have started to have some impact via the expectation channel. - Statistical filters indicate a modest undervaluation (about 9–10 percent), while econometric models (VECM and OLS) point to prices being close to equilibrium by 2015Q2; MRSM indicates boom probability rose in 2013–14 but began to fall in 2015. *Source: IMF staff calculations and cited data in the content unit (_cr16317).* ### 51. The deviation from a long-term trend of the price-to-rent ratio suggests that the CRE ### 51. The deviation from a long-term trend of the price-to-rent ratio suggests that the CRE ### CRE valuation: non-parametric indicators - Price-to-rent ratio suggests the CRE sector was moderately overvalued as of 2015Q3. - The metric shows CRE prices were exuberant before the crisis, growing significantly above rental yield; the post-crisis adjustment was larger than in the RRE market. - From 2014 the ratio breached the historical average again. - Note: the absolute level of overvaluation depends on the choice of the period over which the historical average is calculated. - Text-figure statistical filter results: - Band-pass filter (8–20 year interval): as of 2015Q3, CRE prices were close to the equilibrium level in the range of +2 percent. - Frequency filter (8–25 year interval): indicates 8 percent of undervaluation. - HP filters (one-sided and two-sided, lambda=400,000): Irish CRE prices are estimated to be close to the trend. ### CRE valuation: parametric (econometric) methods - Error correction models (quarterly data 1990Q1–2015Q3): - Long-run equation variables: GDP, credit availability, long-term interest rates. - Short-run equation explanatory variables: estimated residual from long-run equation (or price-to-rent ratio), change in interest rates, growth rate of GDP and credit. - Long-run equation R-squared: 74 percent. - Short-run equation R-squared: 30 percent. - FDI inflows were tested but found statistically insignificant in the long-run equation. - Results: - Econometric models indicate CRE prices are marginally undervalued in the range of 5–8 percent (as of 2015Q3). - Alternative specification using price-to-rent ratio as a pseudo error-correction term shows a large boom-bust cycle; the post-crisis negative deviation closed from 40 to 5–8 percent in a two-year horizon by 2015Q3. ### CRE valuation: Markov regime switching models (MRSM) - MRSMs suggest CRE markets entered into a high regime probability in the second half of 2013 — an early warning signal of another prolonged boom. - Two-regime specification (high or low): - Latest boom-bust cycle lasted 20 years (1993–2013). - Expected duration of the high regime estimated to be over nine years (33 quarters). - Boom regime occurred about 65 percent of the sample period 1990–2015 (longer than the bust regime). - Three-regime specification (high, normal, or low): - Captures dynamics better than two-regime model. - Detects a temporary slowdown between two high growth periods during 1993–2007. - Hints at a recent “pick-up” of CRE prices in recent years, an early warning signal of a prolonged boom. ### Summary of mixed signals and implications for monitoring - Mixed signals across methods: - Error correction models → marginal undervaluation (5–8 percent). - Price-to-rent ratio and MRSMs → early warning of an incipient new boom period. - Conclusion: - Mixed results point to the possibility of an emerging risk and the need for vigilance. - Recommended actions: monitor the market, assess the need for action, close data gaps, review the available toolkit, and be ready for immediate policy actions. ### Policy implications and recommendations (CRE) - Ongoing intensive monitoring of CRE lending by the Central Bank of Ireland is welcome. - Authorities should: - Enhance data collection and continue to allocate sufficient resources for CRE market analyses. - Note the announcement that the Central Bank of Ireland and NAMA will co-fund the development of a CRE statistical system by 2018, to be maintained by the CSO, providing detailed information on sales and lease transactions and construction activities (permissions, commencements, completions). ### Household sector: recent developments and vulnerabilities - Balance sheet and indebtedness trends: - Household financial assets increased by 20 percent from €294 billion to €355 billion during 2008Q3–2015Q4. - Household debt fell by 27 percent from €204 billion (previous peak) to €150 billion over the same period. - Household sector became a net lender in 2009 and has kept this status since then. - Household debt outstanding at €150 billion in 2015Q4 (155 percent of household disposable income and 70 percent of nominal GDP). - Household debt-to-disposable income ratio decreased from 215 percent in 2011Q1 to 155 percent in 2015Q4. - Debt as a proportion of total assets decreased from 28 percent in 2012Q1 to 19 percent in 2015Q4. - Composition and trends: - Mortgage loans account for almost 90 percent of total household loans. - Growth rate (q-o-q) of mortgage loans during 2009Q2–15Q3: minus 1.1 percent on average. - Growth rate (q-o-q) of consumer loans in same period: -3.3 percent. - Mortgage interest-rate exposure (as of 2015Q3 / 2015Q4): - Rate types market share, as of 2015Q3: - PDH: Standard Variable Rate 34.1; Tracker 36.7; Fixed less than 1 year 1.9; Fixed over 1 year 7.3; Sum 80.0. - BTL: Standard Variable Rate 6.0; Tracker 13.7; Fixed less than 1 year 0.0; Fixed over 1 year 0.2; Sum 20.0. - Interest rates, as of 2015Q4: - PDH: Standard Variable Rate 3.96; Tracker 1.07; Fixed 1 to 3 years 3.83; Fixed Over 3 year 3.79. - BTL: Standard Variable Rate 4.61; Tracker 1.1; Fixed 1 to 3 years 4.68; Fixed Over 3 year 4.29. - About 91 percent of mortgage loans have variable interest rates. - Tracker loans and SVR loans accounted for 50.4 percent and 40.1 percent of total mortgage loans, respectively; share of loans with fixed rates over 1 year was 7.5 percent. - Average interest rate on SVR loans at 2015Q4: 3.96 percent (PDH). - Interest rates on existing tracker mortgages are about 1 percent. - Negative equity and arrears: - As of December 2015: - Fifteen percent of Primary Dwelling House (PDH) mortgages remained in negative equity. - 23 percent of Buy to Let (BTL) mortgages remained in negative equity. - Breakdown end-2015: - PDH: 10 percent in negative equity and performing; 5 percent in negative equity and non-performing. - BTL: 12 percent in negative equity and performing; 11 percent in negative equity and non-performing. - NPL ratio declined from post-crisis peak 25.6 percent in 2014Q1 to 20.2 percent in 2015Q4. - Number and volume of mortgage arrears fell by about 36 percent and 35 percent since 2013Q3, respectively; but share of arrears >720 days-past-due rose to about 85 percent of total arrears as of 2015Q4. - Distributional and policy-relevance points: - 43 percent of Irish households do not hold any debt. - Debt burden varies substantially across household segments; indebtedness is a particularly heavy burden for the 34–45 age group. - Detailed loan-level information is useful to design targeted policy measures and to identify vulnerable segments. *Source: IMF staff calculation and Central Bank of Ireland data as presented in the content unit.* ### 70. In this section, the Central Bank of Ireland’s internal Loan Loss Forecasting (LLF) models ### 70. In this section, the Central Bank of Ireland’s internal Loan Loss Forecasting (LLF) models ### Sample and model overview - Sample size: 515,210 mortgage loans in Ireland, totaling €78 billion and 68.4 percent of total mortgage loans at end-2014. - Sample composition highlights (December 2014): - Geography: Non-Dublin 72.3 percent; Dublin 27.7 percent. - Loan purpose: PDH 84.6 percent; BTL 15.4 percent. - Buyer status: Previous Owner 60.4 percent; FTB 39.6 percent. - Age distribution: Under 35 9.6 percent; 35–45 37.7 percent; 45–55 31.5 percent; 55–65 16.3 percent; 65–75 4.0 percent; 75+ 1.0 percent. - Current LTV distribution: Under 60 45.0 percent; 60–70 8.3 percent; 70–80 8.5 percent; 80–90 8.7 percent; 90–100 6.7 percent; 100–120 10.9 percent; 120–150 9.4 percent; 150+ 2.5 percent. - Interest rate type: Fixed 7.3 percent; SVR 51.1 percent; Tracker 41.7 percent. - Model usage and scope: - LLF models used since 2011 for stress tests and research. - Analysis here uses only the PD module (Multi-State Model) applied to loan-level variables: bank, interest rate, current LTV ratio, regional unemployment, interest rate type, BTL status, and loan age (and its natural logarithm). - PD and Probability of Cure (PC) are calculated over a three-year horizon from December 2014, using actual/estimated 2015 values and scenario values for 2016 and 2017. ### Backward-looking vulnerability (opening default stocks at end-2014) - All default rates are on an unweighted count basis. - Key default-stock statistics (percent, at end-2014): - Total: All Loans 14.6; PDH Only 12.4. - Non-Dublin: All Loans 15.2; PDH Only 13.0. - Dublin: All Loans 13.1; PDH Only 11.0. - PDH: 12.4 (PDH Only n/a). - BTL: 26.4 (PDH Only n/a). - Previous Owner: All Loans 14.5; PDH Only 13.4. - FTB: All Loans 11.3; PDH Only 11.2. - By age (All Loans / PDH Only): Under 35 10.0 / 9.9; 35–45 12.1 / 11.5; 45–55 14.0 / 12.9; 55–65 14.5 / 13.6; 65–75 17.2 / 16.6; 75+ 10.6 / 8.7. - By current LTV (All Loans / PDH Only): Under 60 7.5 / 7.0; 60–70 10.3 / 9.3; 70–80 11.5 / 10.1; 80–90 12.9 / 10.8; 90–100 16.9 / 14.6; 100–120 22.8 / 19.6; 120–150 31.2 / 26.7; 150+ 68.1 / 70.0. - By interest rate type (All Loans / PDH Only): Fixed 5.1 / 4.6; SVR 15.8 / 14.0; Tracker 14.7 / 11.9. - Observed patterns: - Small differences across geography and FTB status; higher-income households less likely to default (particularly excluding BTL). - BTL default stock double PDH (26.4 percent vs. 12.4 percent). - Default stocks increase with borrower age (except 75+ small sample); 65–75 has 16.6 percent (PDH Only), under 35 has 9.9 percent (PDH Only). - In PDH segment, LTV under 90 typically 7–11 percent default; negative equity categories show default stocks rising: 100–120 LTV 20–27 percent range, 120–150 higher, 150+ much higher (up to 70 percent). - Fixed-rate loans much lower default propensity (4.6 percent PDH Only) than variable-rate mortgages (SVR 14.0 percent PDH Only; tracker 11.9 percent PDH Only). ### Forward-looking vulnerability under FSAP macroeconomic scenarios - Scenario inputs (selected macro indicators; 2015 (est) and 2016–2018 values reported): - Real GDP growth: 2015 (est) 5.8; 2016 4.7; 2017 4.0; Baseline 2018 3.0; Adverse 2016 1.1; Adverse 2017 0.3; Adverse 2018 0.1. - Unemployment rate: 2015 (est) 9.5; 2016 8.5; 2017 7.8; Baseline 2018 7.2; Adverse 2016 9.6; Adverse 2017 10.5; Adverse 2018 11.2. - Short-term interest rate (3 month Euribor): 2015 (est) 0.0; 2016 0.0; 2017 0.1; Baseline 2018 0.4; Adverse 2016 0.8; Adverse 2017 0.9; Adverse 2018 1.2. - Long-term interest rate (10 year Irish government bond): 2015 (est) 1.3; 2016 1.7; 2017 2.3; Baseline 2018 3.3; Adverse 2016 3.2; Adverse 2017 5.3; Adverse 2018 7.9. - HICP growth: 2015 (est) 0.3; 2016 1.5; 2017 1.9; Baseline 2018 1.6; Adverse 2016 0.5; Adverse 2017 0.3; Adverse 2018 -0.7. - Residential property prices: 2015 (est) 6.5; 2016 7.5; 2017 6.7; Baseline 2018 5.1; Adverse 2016 -5.0; Adverse 2017 -5.0; Adverse 2018 0.0. - Commercial property prices (selected): 2015 (est) 21.8; 2016 10.3; 2017 4.2; Baseline 2018 1.9; Adverse 2016 -4.6; Adverse 2017 -16.1; Adverse 2018 -11.2. - PD calculation method: - Three-year PDs for performing loans at December 2014 are calculated by sequentially applying year 1 PD and PC to derive the probabilistic default through end-2017, incorporating scenario-driven changes to interest rates, regional unemployment, and loan-level LTV. - PDs reported are unconditional average forward-looking vulnerabilities among loans that were performing at end-2014. - Key forward-looking results: - LTV and stressed PDs: strictly increasing relationship; LTV over 150 percent shows adverse PDs climbing above 20 percent, with a significant share closer to 30 percent. - Dublin vs Non-Dublin (average three-year PD, 2014–17): - Baseline: close to 6 percent. - Adverse: close to 8 percent. - Loans outside Dublin only marginally more vulnerable than loans in Dublin. - PDH vs BTL (average three-year PD, 2014–17): - Forward-looking unconditional average PDs show a smaller difference between PDH and BTL than backward-looking default stocks, because the most vulnerable BTL loans had already defaulted before December 2014. - Compositional factors: performing BTLs have a higher prevalence of tracker mortgages (63 percent BTL vs. 38 percent PDH), leading to lower interest rates among performing BTLs. - FTB vs Previous Owner (average three-year PD, 2014–17): - Performing FTBs: average adverse three-year PD over 9 percent. - Previous owners: average adverse three-year PD closer to 7 percent. - Differences driven by current LTV distribution among performing loans: average CLTV 53 for previous owners vs. 75 for FTBs; 90th percentile CLTV 103 for previous owners vs. 125 for FTBs. - Age groups (average three-year PD, 2014–17): - Under 35: adverse PD 10 percent on average. - 65–75: adverse PD 6 percent on average. - Younger performing borrowers have higher average and 90th percentile LTVs (e.g., under 35 average LTV 83 and 90th percentile 128) than older borrowers (65–75 average LTV 41 and 90th percentile 106). - Interest rate type (average three-year PD, 2014–17): - Tracker loans: projected PD 6 percent under the adverse FSAP scenario. - SVR loans: projected PD closer to 10 percent under the adverse FSAP scenario. - Fixed-rate loans have lower forward-looking PDs relative to SVR. ### Summary of findings - Backward-looking default risk (stock at end-2014) is extremely high among: - BTL loans (26.4 percent). - Loans in deep negative equity (e.g., current LTV 120–150: 31.2 percent; 150+: 68.1 percent). - Variable-rate mortgages (SVR and tracker) are significantly more impaired than fixed-rate loans. - Older borrowers and previous-owner mortgage holders exhibit higher default stocks than FTBs. - Forward-looking vulnerabilities under the adverse scenario identify at-risk segments: - First-time buyers. - Younger borrowers. - Borrowers on SVR mortgages (predicted rate closer to 10 percent). - Borrowers outside Dublin (to a lesser extent). - Tracker mortgages are estimated to have a three-year PD of 6 percent under the adverse FSAP scenario. - Distinction emphasized: - The largest source of mortgage credit risk in most household segments is the existing stock of defaulted loans rather than the probabilistic default risk among performing loans. ### Policy implications - The analysis supports the role of proportionate limits on LTV and LTI ratios. - Recommendation: Central Bank of Ireland should maintain the present state of the measures to build resilience of borrowers and lenders against future macro-financial shocks. - Rationale: - LTI limits reduce the probability of defaults. - LTV limits bolster borrowers’ resilience to house price shocks by increasing equity in residential property and reduce loss given default for lenders. - LTV caps without LTI limits could leave borrowers vulnerable to income shocks; LTI caps without LTV measures could leave lenders highly exposed to severe house price shocks (as occurred in Ireland after 2008). - Monitoring: Closely monitor the risky segments identified (high LTV ratios, young borrowers, FTBs, BTL investors, and SVR borrowers). *Sources: Central Bank of Ireland; CBI in collaboration with the ECB; IMF staff calculations.* ### 88. It will be crucial to continue to collect loan-level data from commercial banks for systemic ### _cr16317 - 88. It will be crucial to continue to collect loan-level data from commercial banks for systemic ### Loan-level data, Central Credit Register, and macroprudential monitoring - It will be crucial to continue to collect loan-level data from commercial banks for systemic risk assessment and macroprudential implementation. - The Central Bank of Ireland could consider expanding loan-level data collection to all regulated financial institutions for monitoring if the number of institutions lending in the market increases. - The loan-level data allowed the Central Bank of Ireland to gather granular information on the Irish mortgage market previously inaccessible with aggregate data. - The establishment of a Central Credit Register (CCR) will further support the gains in data granularity and risk identification achieved to date. ### Corporate sector recent performance and headline statistics - 2015 corporate sector real growth rate: 12 percent. - 2015 gross operating surplus increased by 15 percent in real terms. - Profitability share (gross operating surplus divided by gross value added) reached about 65 percent versus a pre-crisis average of 55 percent. - Non-financial corporate (NFC) sector debt: 187 percent of GDP (2015Q4). - NFC debt-to-equity ratio: 45 percent in 2015Q4, down from a peak of 123 percent in early 2009. - Multinationals’ consolidated debt: 94½ percent of GDP in 2014 compared to 36 percent of GDP in 2007. - Debt of Irish firms: 85 percent of GDP in 2014 from about 65 percent of GDP in 2007. - Share of loans from non-Irish entities: 55 percent of total loans in 2015Q4 compared with 34 percent at end-2012. - Share of non-euro denominated debt in total debt increased to more than half in 2015 from about one-quarter in 2006. ### Vulnerabilities, non-performing loans, and default indicators - Corporate sector non-performing loans (NPLs): about 17 percent in 2015Q3 from a peak of nearly 27 percent at end-2013. - NPL ratio in construction sector: stayed above 50 percent (despite moderation). - Corporate bankruptcies: quarterly average about 300 (2009–2012) versus 84 (2005–07); more recently quarterly average about 225 cases (2014–15). - Black-Scholes-Merton (BSM) and Altman Z-probability: - 2014: BSM 4½ percent; Z-probability 14 percent. - 2008: BSM 24 percent; Z-probability 32 percent. - Distance-to-default (DTD) improved significantly across main sectors; in 2014 DTD of Irish corporate sector slightly better than 2014 euro zone average. ### Firm-level data, sample, and methodology - Data source: ORBIS database (Bureau Van Dijck). - Sample coverage: - Unbalanced panel dataset of 7,663 observations of 3,040 domestic firms. - 1,904 observations of 727 foreign firms. - Total: 9,567 observations and 3,767 firms. - Firm size composition (domestic): - Small (1-49 employees): 4,409 observations; 1,882 firms. - Medium (50-249 employees): 2,266 observations; 833 firms. - Large (>249 employees): 988 observations; 325 firms. - Firm size composition (foreign): - Small (1-49 employees): 1,131 observations; 475 firms. - Medium (50-249 employees): 543 observations; 188 firms. - Large (>249 employees): 230 observations; 64 firms. - Sectoral composition (number of observations): - Primary: 156 total (121 domestic; 35 foreign). - Manufacturing: 1,294 total (999 domestic; 295 foreign). - Construction: 734 total (603 domestic; 131 foreign). - Services: 7,383 total (5,940 domestic; 1,443 foreign). - Sample period: 1995–2014, with majority of observations concentrated in 2006–2013. - The analysis focuses on interest cover ratio (ICR = EBITDA / interest expenses); ICR < 1 indicates technical default; ICR between 1 and 2 is regarded as being at heightened risk. ### Limitations of the firm-level dataset - Composition and number of firms are not fixed over time, complicating inter-temporal comparisons. - Potential selection bias: distressed firms may be under-represented. - Possible sectoral under- or over-representation. - Small firms appear under-represented relative to their share in the Irish economy. - Mitigations: vulnerability indicators and sensitivity analysis are presented by firm size; impact on banks’ asset quality and capital position also reported using a sample adjusted for the “right” composition of firms. ### Firm-level stylized facts and trends - Median return on equity (RoE): fell to just below 3 percent in 2009 from 14 percent in 2007. - Profitability recovery in 2010–13 was modest across firm sizes and sectors; RoE of large firms remained significantly above that of smaller firms. - Debt-to-equity, effective interest costs, and sectoral compositions are analyzed (figures referenced for trends by sector and firm size). *Source: Excerpt from IMF technical note and corporate sector analysis in the 2016 Ireland Financial Sector Assessment Program* ### 105. The debt-to-equity ratio has declined since the pre-crisis period, though deleveraging was ### _cr16317 - 105. The debt-to-equity ratio has declined since the pre-crisis period, though deleveraging was ### Leverage and debt-to-equity - The debt-to-equity ratio of the median firm in the sample declined to 52 percent in 2013 from 65 percent in 2007, driven largely by a rapid deleveraging of small firms. - The debt-to-equity ratio of the median large firm increased to 100 percent in 2013 from 75 percent in 2007. - Deleveraging was evident across all sectors, though at a slower pace among firms in the services sector. ### Interest costs and Interest Cover Ratio (ICR) dynamics - Median firm effective interest cost: - 2007: 4.6 percent (previous year) - 2008: increased to 5 percent - Following years: eased to around 3.5 percent. - Effective interest rate cost of large firms was higher than that of smaller firms (possible reasons: longer maturities and higher term premia). - ICR developments: - Median firm ICR declined to about 4 in 2008–09 from just above 6 in 2007. - Share of vulnerable domestic firms (ICR < 2): - 2007: 24.4 percent - 2009: 37.2 percent - Share of domestic firms with ICR < 1 also increased significantly in 2008–09. - Recovery by 2013: - ICR of the median domestic firm in 2013: 6.3 - ICR of the median foreign firm in 2013: 6.6 - Small firms’ ICR remained below that of medium-sized and large firms in both domestic and foreign ownership. ### Sectoral patterns of ICR - U-shape pattern across sectors: median ICR fell in all main sectors in 2008–09 and improved thereafter. - Primary sector: - ICR of the median firm fell to below 1 in 2009 from above 10 in 2006; half of firms in “technical default”. - Construction sector: - Sharp decline in median ICR in 2008–09, but financial health of the median firm remained solid. - By 2013 the ICR of the median firm was above the 2007 level across all sectors. ### Debt-at-risk and composition by firm size and ownership - Share of debt owned by domestic firms with ICR < 1: - 2007: 10.5 percent - 2009: above 15 percent - 2013: 9.9 percent (reverted to just below pre-crisis level with different composition) - Composition changes (domestic firms, share of risky debt in 2013): - Small firms: about two-thirds of risky debt (compared with one-third in 2007) - Large firms: declined to about 15 percent of risky debt in 2013 (from nearly 50 percent in 2007) - Share of debt owned by foreign firms with ICR < 1: - 2007: 11.2 percent - 2009: 20 percent - 2013: 4.1 percent - Table 12 (Share of Vulnerable and Distressed Firms, percent): - Domestic firms: ICR<2 — 2007: 24.4; 2009: 37.2; 2013: 24.3 - Domestic firms: ICR<1 — 2007: 16.6; 2009: 27.7; 2013: 18.0 - Foreign firms: ICR<2 — 2007: 28.4; 2009: 35.2; 2013: 19.1 - Foreign firms: ICR<1 — 2007: 20.4; 2009: 27.7; 2013: 12.7 - (Table breaks down shares also by Small, Medium, Large within Domestic and Foreign categories as presented in source.) - Table 13 (Debt-at-Risk, percent of total debt held by firms with ICR < 1): - Domestic firms: 2007: 10.5; 2009: 15.3; 2013: 9.9 - Domestic small: 2007: 3.1; 2009: 6.5; 2013: 6.3 - Domestic medium: 2007: 2.3; 2009: 4.2; 2013: 2.3 - Domestic large: 2007: 5.1; 2009: 4.6; 2013: 1.3 - Foreign firms: 2007: 11.2; 2009: 20.0; 2013: 4.1 - Foreign small: 2007: 5.2; 2009: 4.9; 2013: 3.0 - Foreign medium: 2007: 2.4; 2009: 3.6; 2013: 0.8 - Foreign large: 2007: 3.6; 2009: 11.5; 2013: 0.3 ### Econometric analysis of distress (Logit model) - Objective: test relationship between low ICR (below 2) and probability of being in “distress” (insolvency proceedings, liquidation, default of payment, or temporary inactivity). - Model: binary Logit with dependent variable = probability of distress; explanatory variables include firm-level factors, sectoral and time dummies; two alternative dummies for ICR thresholds (ICR < 2 and ICR < 1). - Sample: 1995–2014, about 9,500 observations; 706 observations of distress (~7½ percent), with more than half in 2007–09. - Key econometric findings: - ICR is an important indicator of corporate financial soundness; firms with low ICR are more likely to be in distress. - Marginal effects at mean indicate the vulnerable dummy (ICR < 2) effect is slightly higher than the highly vulnerable (ICR < 1) dummy in some specifications. - Firms in the construction sector are more likely to be in distress; some specifications show manufacturing firms also more prone to distress. - Higher current ratio (more liquid companies) reduces probability of distress. - Larger companies are less likely to be in distress. - Ownership (domestic vs. foreign) did not show a significant effect in the reported specifications. - Table 14 (selected Logit coefficients and model stats): - Highly vulnerable coefficients (columns with this dummy): 0.806*, 0.804*, 0.795* - Vulnerable coefficients (columns with this dummy): 0.948*, 0.956*, 0.945* - Current ratio coefficients: examples include -0.024**, -0.032**, -0.032**, -0.022*, -0.031*, -0.031** - Fixed assets/total assets: coefficients around -0.004* to -0.005* - Number of employees: -0.000** (significant) - Manufacturing: coefficients around 0.189, 0.144, 0.168 and significance in some specs (0.224***, 0.174, 0.196***) - Construction: 0.815*, 0.731*, 0.740*, 0.824*, 0.730*, 0.738* - Observations: 9,522 for all reported specifications - Pseudo R2: 0.086 to 0.100 - Significance notation: * 1 percent; ** 5 percent; *** 10 percent. ### Sensitivity analysis: calibration of shocks - Sample: excludes subsidiaries of foreign firms; static exercise based on end-2013 balance sheets. - Three shocks assessed: - Interest rate shock: - Apply a 450bp shock to the effective interest rate of each firm at end-2013. - Assume half of the 2013 debt stock will be rolled-over with a higher interest rate. - Add expected increase in return on financial assets to EBITDA, assuming a 2-percentage points spread between lending rates and deposit rates will prevail; yield on financial assets assumed to increase by 250bp. - Profit shock: - Shock firms’ added value by 15 percent (holding costs of employees constant). - Calculation implies average profit declines by about 20 percent compared with the baseline. - Combined interest rate and profit shock: applies both effects to numerator and denominator as defined in source formulas. - Assumptions and notes: - Rollover assumption of 50 percent accounts for higher weight of large firms (60 percent) in total debt. - Firms’ financial assets assumed to amount to 10 percent of firms’ total assets. - Parameters aligned with FSAP’s adverse scenarios and consistent with banking-sector stress test deviations. ### Stress test results (Table 15) — Corporate sector sensitivity analysis results (percent except where indicated) - Baseline (All firms): - ICR of the median: 6.8 - Share of firms ICR<1: 16.5 - Share of firms ICR<2: 22.3 - Share of debt ICR<1: 9.4 - Share of debt ICR<2: 22.7 - Baseline by size: - Small: median ICR 6.0; Share of firms ICR<1 8.0; ICR<2 11.5; Share of debt ICR<1 5.8; ICR<2 6.1 - Medium: median ICR 7.4; Share of firms ICR<1 6.9; ICR<2 8.8; Share of debt ICR<1 2.3; ICR<2 9.6 - Large: median ICR 7.4; Share of firms ICR<1 1.5; ICR<2 2.2; Share of debt ICR<1 1.3; ICR<2 7.0 - Interest rate shock (450bp) — All firms: - ICR of the median: 4.0 - Share of firms ICR<1: 17.3 - Share of firms ICR<2: 31.0 - Share of debt ICR<1: 9.8 - Share of debt ICR<2: 56.6 - Interest rate shock by size (selected): - Small: median ICR 3.6; Share of firms ICR<2 17.2; Share of debt ICR<2 7.0 - Medium: median ICR 4.3; Share of firms ICR<2 10.8; Share of debt ICR<2 21.6 - Large: median ICR 4.7; Share of firms ICR<2 3.0; Share of debt ICR<2 27.0 - Profit shock — All firms: - ICR of the median: 2.0 - Share of firms ICR<1: 42.9 - Share of firms ICR<2: 50.2 - Share of debt ICR<1: 11.3 - Share of debt ICR<2: 25.9 - Profit shock by size (selected): - Small: median ICR 1.9; Share of firms ICR<1 24.7; Share of debt ICR<2 7.1 - Medium: median ICR 2.1; Share of firms ICR<1 14.5; Share of debt ICR<2 10.1 - Large: median ICR 2.3; Share of firms ICR<1 3.7; Share of debt ICR<2 8.7 - Combined shock — All firms: - ICR of the median: 1.2 - Share of firms ICR<1: 47.4 - Share of firms ICR<2: 58.6 - Share of debt ICR<1: 14.7 - Share of debt ICR<2: 59.9 - Combined shock by size (selected): - Small: median ICR 1.2; Share of firms ICR<1 26.4; Share of debt ICR<2 7.6 - Medium: median ICR 1.2; Share of firms ICR<1 16.5; Share of debt ICR<2 22.5 - Large: median ICR 1.3; Share of firms ICR<1 4.5; Share of debt ICR<2 29.8 ### Interpretation of stress results and implications - A 450bp interest rate shock: - Would push median ICR from 6.8 to 4.0 (similar to 2008–09 median). - Would increase share of vulnerable firms (ICR < 2) to nearly one-third from 22 percent baseline. - Would increase share of risky debt (debt owned by firms with ICR < 2) to more than half of total (56.6 percent), signaling potential for sharp increase in new defaults. - Most vulnerable groups by debt share under this shock are large and medium-sized firms, reflecting high leverage and debt service payments. - A profit shock: - Would bring median ICR to 2.0 and make more firms vulnerable compared with the interest rate shock (higher share of firms with ICR < 2). - Impact on share of debt owed by firms with ICR < 2 would be significantly smaller than under interest rate shock (25.9 percent vs. 56.6 percent), suggesting interest rate shock disproportionately affects heavily leveraged firms. - Sample evidence: firms with ICR < 2 under profit shock had average debt-to-equity 140 percent in baseline; under interest rate shock average debt-to-equity 197 percent in baseline. - Combined shock: - Would increase share of firms with ICR < 2 to nearly 60 percent from 22 percent baseline. - Share of risky debt would be at 50 percent or above across all firm-size categories (Table 3 in Appendix IV), indicating sharp increase in potential defaults. - Implications for banks: - The analysis proceeds to estimate banks’ potential losses from corporate exposures and compare with banks’ buffers using methodology aligned with the GFSR (IMF 2013). *Source: BvD and IMF staff’s estimates.* ### Appendix IV). ### Appendix IV ### Methodology and Key Assumptions - Loss rates are obtained by multiplying the PDs by loss given default (LGDs) ratios; Basel’s standard LGDs of 45 percent are applied. - Assumed shares of firm debt owned by banks: - 50 percent of the large firms’ debt is owned by the banks. - 75 percent of the SMEs’ debt is owned by the banks. - Banks’ corporate loan book is assumed to remain constant during the shock period. - Calculations are based on Moody’s PDs; calculations based on alternative PDs are not significantly different. - The impact analysis applies the scenario-specific share of new defaults in the total corporate loan book to the aggregate corporate loan book of the three main domestic banks in 2015Q3. - Corporate loans account for about 35 percent of the domestic banks’ loan book. ### Results — New Corporate Defaults (cumulative, two-year horizon) - Baseline new corporate defaults: - 3.6 percent of the overall bank corporate loan book (baseline used in main estimation). - Profit shock: - Increases new corporate defaults to 5.7 percent of the overall bank corporate loan book from 3.6 percent in the baseline. - Interest rate shock: - Increases new corporate defaults to 7.1 percent of total corporate loans. - Combined profit-interest rate shock: - Increases new corporate defaults to 9.4 percent of total corporate loans. - Comparison: - These estimates appear comparable to the 11 percentage point increase in corporate NPL ratio realized in 2011–13. ### Sensitivity to Ownership and LGD Assumptions - If the share of SMEs’ debt owned by banks is assumed equal to large firms (50 percent): - Baseline: 3.2 percent (cumulative, two-year horizon). - Profit shock: 5.0 percent of total corporate loans. - Interest rate shock: 6.6 percent of total corporate loans. - Combined shock: 8.9 percent of total corporate loans. - If a more conservative LGD of 60 percent is used: - Results are not dramatically different in default-share estimates. - Impact on banks’ capital (see next section) increases loss severity. ### Impact on Banks’ Regulatory Tier 1 Capital - Baseline regulatory Tier 1 capital (2015Q3): 17.4 percent of risk-weighted assets. - Combined shock (LGD = 45 percent): - Regulatory Tier 1 capital would fall to 15.1 percent of risk-weighted assets from 17.4 percent in the baseline (2015Q3). - Combined shock (LGD = 60 percent): - Regulatory Tier 1 capital would fall by an additional 0.8 percentage points of risk-weighted assets to 14.3 percent. - Re-weighting sample toward higher SME representation (see next subsection) increases capital impact: - Baseline (re-weighted sample): 5.2 percent new defaults (cumulative, two-year horizon). - Profit shock: 9.1 percent of total corporate loans. - Interest rate shock: 9.4 percent of total corporate loans. - Combined shock: 11.9 percent of total corporate loans. - In the combined shock under the re-weighted sample, regulatory Tier 1 capital would fall to 14.5 percent of risk-weighted assets from 17.4 percent in the baseline (2015Q3). ### Re-weighting the Composition of Firms - Rationale: - The baseline sample (observations of 2013) under-represents SMEs: sample SMEs account for about 90 percent of total firms versus a Central Statistics Office estimation of 99.7 percent. - SMEs have higher reliance on bank financing and weaker initial financial health. - Effect of increasing SME weight while keeping ICR distribution unchanged: - Shocks produce a stronger effect on new defaults and banks’ capital (see figures above). ### Conclusions and Caveats - Recent developments: - Corporate sector vulnerabilities moderated recently: increased profitability, sharp reduction in debt-to-equity ratio, decline in distress indicators, and an increase in the median firm ICR since 2008–09. - NFC debt-to-GDP ratio remains high from an international perspective, but more than half is owned by multinationals and originates from external sources, including inter-group loans. - Remaining vulnerabilities: - Corporate debt and corporate NPLs remain high. - About one-fifth of domestic firms were under “technical default” (ICR < 1) in 2013; the share of debt owned by firms with ICR < 1 was 10 percent. - The share of risky debt among small firms constituted nearly half of small firms’ debt. - Interpretation and limitations: - The sensitivity analysis suggests adverse shocks would push many firms into a vulnerable state and substantially increase new defaults, yet banks’ regulatory Tier 1 capital would remain above the minimum requirement in this first-round exercise. - The analysis is static and does not account for second-round effects (e.g., higher unemployment, lower property prices) that could amplify losses on banks’ balance sheets. - Adverse macroeconomic changes could have wider impacts on government revenues, households, and financial institutions, implying the stress on banks’ balance sheets is likely to be greater than measured here. *Appendix IV, IMF staff calculations and text as provided in the source document.* ### Appendix Table 2. Ireland: Share of Vulnerable and Distressed Firms /1 ### Appendix Table 2. Ireland: Share of Vulnerable and Distressed Firms /1 ### Share of vulnerable and distressed firms (Share of firms per year, percent) - 2007 - Domestic firms: ICR<2 24.4; ICR<1 16.6 - Within small firms: ICR<2 26.9; ICR<1 18.9 - Within medium firms: ICR<2 20.7; ICR<1 12.9 - Within large firms: ICR<2 22.2; ICR<1 15.7 - Foreign Firms: ICR<2 28.4; ICR<1 20.4 - Within small firms: ICR<2 30.3; ICR<1 20.2 - Within medium firms: ICR<2 24.0; ICR<1 19.8 - Within large firms: ICR<2 30.3; ICR<1 24.2 - 2009 - Domestic firms: ICR<2 37.2; ICR<1 27.7 - Within small firms: ICR<2 37.8; ICR<1 28.9 - Within medium firms: ICR<2 36.5; ICR<1 27.0 - Within large firms: ICR<2 36.0; ICR<1 21.6 - Foreign Firms: ICR<2 35.2; ICR<1 27.7 - Within small firms: ICR<2 35.6; ICR<1 26.6 - Within medium firms: ICR<2 33.3; ICR<1 26.9 - Within large firms: ICR<2 38.5; ICR<1 38.5 - 2013 - Domestic firms: ICR<2 24.3; ICR<1 18.0 - Within small firms: ICR<2 24.9; ICR<1 17.8 - Within medium firms: ICR<2 24.4; ICR<1 19.5 - Within large firms: ICR<2 20.4; ICR<1 14.3 - Foreign Firms: ICR<2 19.1; ICR<1 12.7 - Within small firms: ICR<2 17.7; ICR<1 12.9 - Within medium firms: ICR<2 20.6; ICR<1 11.8 - Within large firms: ICR<2 21.4; ICR<1 14.3 - Note: 1/ The number of firms changes over time. ### Debt-at-Risk (Debt of firms with ICR < 1 relative to total debt, percent) - 2007 - Domestic firms: 10.5 - Within small firms: 22.2 - Within medium firms: 8.0 - Within large firms: 8.8 - Foreign firms: 11.2 - Within small firms: 28.2 - Within medium firms: 21.2 - Within large firms: 8.9 - 2009 - Domestic firms: 15.3 - Within small firms: 40.4 - Within medium firms: 20.7 - Within large firms: 7.2 - Foreign firms: 20.0 - Within small firms: 23.2 - Within medium firms: 21.7 - Within large firms: 41.3 - 2013 - Domestic firms: 9.9 - Within small firms: 49.0 - Within medium firms: 8.9 - Within large firms: 2.1 - Foreign firms: 4.1 - Within small firms: 38.6 - Within medium firms: 7.8 - Within large firms: 0.6 ### Determinants of Firms’ Distress (Logit model, Marginal effects at the means) - Model sample and fit - # of Obs.: 9,522 (for models (1) through (6)) - Pseudo R2 by model: - (1) 0.086 - (2) 0.088 - (3) 0.097 - (4) 0.0957 - (5) 0.098 - (6) 0.100 - Time dummies: yes (in all models) - Significance notation: * Indicates significance at 1 percent; ** indicates significance at 5 percent; *** indicates significance at 10 percent. - Reported marginal effects (selected coefficients and significance preserved as in the source) - Highly vulnerable: 0.038* (models (1), (2), (3)) - Vulnerable: 0.044* (models (4), (5), (6) reporting 0.044*, 0.044*, 0.043*) - Current ratio: -0.001**; -0.001**; -0.001**; -0.001*; -0.001*; -0.001** (models (1)–(6) respective entries) - Fixed assets/total Assets: -0.000* (appears in multiple model columns as -0.000*) - Number of employees: -0.000** (appears in two model columns) - Foreign: 0.004 (reported in model columns where present) - Primary sector: -0.012; -0.010; -0.012; 0.010; -0.009; -0.010 (as reported across models) - Manufacturing: coefficients reported across models including entries 0.009; 0.006; 0.007; 0.009***; 0.008; 0.009*** - Construction: 0.039*; 0.034*; 0.034*; 0.033*; 0.033*; 0.033* (across models) ### Mapping and Sensitivity Analysis Results - Appendix Table 5: Mapping of Corporate Vulnerability Indicators to Probabilities of Default (Source: IMF, GFSR, October 2013, Chapter 1, Annex 1.2.) — mapping referenced in the original content. - Appendix Table 6: Sensitivity Analysis Results (Shares of Firms and Shares of Debt; baseline 2013 and shocks) - Baseline, 2013 - Shares within small firms: ICR<1 14.9; ICR<2 21.2; Share of Debt ICR<1 47.3; ICR<2 49.9 - Shares within medium-sized firms: ICR<1 19.5; ICR<2 24.3; Share of Debt ICR<1 8.9; ICR<2 36.5 - Shares within large firms: ICR<1 14.2; ICR<2 20.4; Share of Debt ICR<1 2.1; ICR<2 11.4 - Interest rate shock - Shares within small firms: ICR<1 16.1; ICR<2 31.7; Share of Debt ICR<1 50.3; ICR<2 57.5 - Shares within medium-sized firms: ICR<1 20.7; ICR<2 30.5; Share of Debt ICR<1 9.0; ICR<2 82.9 - Shares within large firms: ICR<1 12.2; ICR<2 28.6; Share of Debt ICR<1 2.1; ICR<2 44.1 - Profit shock - Shares within small firms: ICR<1 45.8; ICR<2 51.4; Share of Debt ICR<1 48.7; ICR<2 51.6 - Shares within medium-sized firms: ICR<1 40.9; ICR<2 49.4; Share of Debt ICR<1 9.9; ICR<2 38.5 - Shares within large firms: ICR<1 34.7; ICR<2 46.9; Share of Debt ICR<1 3.8; ICR<2 14.1 - Combined shock - Shares within small firms: ICR<1 49.0; ICR<2 58.2; Share of Debt ICR<1 52.2; ICR<2 59.0 - Shares within medium-sized firms: ICR<1 46.3; ICR<2 61.0; Share of Debt ICR<1 12.3; ICR<2 85.7 - Shares within large firms: ICR<1 42.9; ICR<2 53.1; Share of Debt ICR<1 7.9; ICR<2 48.8 ### Distribution of distress by sector and firm size (figures summary) - Sectoral distribution (Figure 4, percent of total observations) - Primary: 1% - Manufacturing: 15% - Construction: 24% - Services: 60% - Firm size distribution (Figure 4, percent of total observations) - Large: 12% - Medium: 33% - Small: 55% _Appendix tables and figures as presented in the source PDF._ --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16317.pdf_