## EXECUTIVE SUMMARY ## Source details **Canonical URL:** [EXECUTIVE SUMMARY](https://www.imf.org/-/media/files/publications/cr/2018/cr18164.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2018/cr18164.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2018/cr18164.pdf.json) --- ### Summary of systemic risk monitoring and macroprudential capacity - The National Bank of Romania (NBR) has long experience in implementing macroprudential policy measures and a relatively sophisticated systemic risk monitoring framework, including: - Monitoring several indicators derived from the nation-wide credit register and related data sources. - Constructing summary indicators to facilitate overall risk assessment. - Using various economic models to assess macro-financial developments, shock impacts, and policy actions. - Regular solvency and liquidity stress tests for banks. - Data and information gaps: - Extensive NPL exposures sold to asset management companies do not have to report to the credit registry, creating key data gaps. ### Institutional framework: mandate, powers, and governance concerns - Institutional arrangements and recent changes: - National Committee for Macroprudential Oversight (NCMO) established April 2017 as the national macroprudential authority with a clear legal mandate. - NCMO is chaired by the Governor of the NBR; Secretariat resides within the NBR. - NCMO has nine members, including three NBR representatives (reduced from a proposed five), giving NBR the same number of representatives as the ASF and the Ministry of Public Finance (MoPF). - Functioning and governance: - As of April 2018, NCMO held five meetings and issued ten recommendations, including quarterly recommendations on the countercyclical capital buffer (CCyB). - Technical Commissions on Systemic Risk and Financial Crisis Management (TCSR and TCFCM) were yet to become operational; composition approved February 2018. - Powers and tools: - NCMO has direct (hard) powers over a wide range of macroprudential tools, power to recommend actions to other authorities or Government with a “comply or explain” mechanism, and power to issue warnings and opinions. - Key institutional concerns and recommended governance reforms: - Identified inaction bias: cost of policy actions is sooner and more observable than their potential benefits. - NBR can propose policy action and be outvoted without disclosure of decision or vote distribution or linkage to a common risk assessment. - Recommended governance reforms: - Require proposed policy actions and distribution of votes to be publicly disclosed in the summary of meetings. - Develop a common assessment of systemic risk at each NCMO meeting to foster consensus and common ownership of actions. - Allow for exemptions to public disclosure where unanimous NCMO vote deems disclosure likely detrimental to financial stability. ### Macroprudential toolkit: current instruments and gaps - Recent framework developments: - CRD/CRR framework operational in Romania; capital buffers and Liquidity Coverage Ratio (LCR) implemented or being phased-in. - Longer-standing borrower tools remain but scope is narrow. - Existing borrower- and sector-specific calibrations (as stated in source): - Maximum LTV on domestic currency-denominated mortgages: "Eighty-five percent for domestic currency-denominated mortgages, except for Prima Casa mortgages." (October 2011) - Maximum LTV on FX mortgages: "Eighty percent for FX loans to hedged borrowers, 75 percent for EUR-denominated mortgages to unhedged borrowers, and 60 percent in other currencies to unhedged borrowers." (October 2011) - Stressed DSTI on consumer loans: "Credit institutions are responsible for testing borrowers’ resilience throughout the duration of the credit, and the maximum indebtedness level should be determined by taking into consideration the interest rate risk, the currency risk, and the risk of a reduction in income." (October 2011) - Other restrictions on consumer loans: "Maturity restricted to five years. Value of goods purchased for consumer credit in FX capped at 133 percent of the credit amount." (October 2011) - Higher risk weights on commercial real estate exposures: "100 percent risk weights on commercial real estate exposures." (January 2014) - Other systemically important institutions (OSII) buffer: "One percent of total risk exposure for nine credit institutions identified as having systemic importance." (January 2018) - Identified toolkit gaps and recommended extensions: - Apply a stressed DSTI limit to all loans (mortgages and consumer loans). - Gradually scale back the Prima Casa program to mitigate housing sector imbalances and support LTV limit effectiveness. - Enforce a currency-differentiated LCR for significant currencies and monitor a currency-differentiated NSFR. - Strengthen oversight and provisioning for nonbank financial lenders (NBFLs), aligning provisioning regimes with banks. - Require asset management companies (debt collecting agencies specializing in NPL purchases) to report to the credit registry. ### Systemic vulnerabilities and policy implications - Main vulnerabilities identified: - Sovereign-bank nexus: - Banks have concentrated and large exposures towards the domestic sovereign; sovereign debt exposure of banks was around 20 percent of assets as of December 2016 (up from below 5 percent in 2008). - Duration of domestic sovereign debt held by banks is long relative to banks’ funding. - Public debt roughly 36.8 percent of GDP at year-end 2017; NBR analysis points to debt levels between 40 and 45 percent of GDP as thresholds for material increase in recession likelihood and refinancing risk. - FSAP adverse scenario: total drop of 300bps in capital over the 3-year horizon due to sovereign exposures; trading gains of 0.6 percent of RWA in 2017 turn to a market loss of 3.6 percent in 2018 for the system as a whole. - Housing and household sector: - Prima Casa: State guarantees 50 percent of the mortgage; down payment 5 percent of property value; interest rate capped at ROBOR plus 2.5 percent. - Housing loans rose from 21 percent of loans to households to more than 58 percent between 2008 and 2016. - Annual house price growth of 6.2 percent as of September 2017. - About 25 percent of debtors had a DSTI level in excess of 55 percent (September 2017). - Majority of mortgage contracts at variable rates; stress tests show real estate mortgage losses account for a majority of overall credit losses in an adverse scenario. - Corporate sector and funding shifts: - Corporate leverage increased between 2012 and 2016 despite a decline in bank credit; more than 40 percent of firms have negative equity; about 30 percent of these companies were established over the last five years. - Corporate external borrowing and reliance on trade credit increased substantially. - FX-related vulnerabilities: - FX loans were 44 percent and 42 percent of total corporate and household loans, respectively, in 2017. - NPLs higher for FX loans than domestic currency: 12.3 percent vs 3.3 percent in December 2016. - Many corporate FX borrowers are unhedged; NBFLs lend predominantly in FX (84 percent). - NBFLs and regulatory arbitrage: - Rapid lending growth to SMEs and low-income households in NBFL sector; NBFL external borrowing and lending pose regulatory arbitrage risks. - Policy implications and recommended measures: - Use the Systemic Risk Buffer (SRB) to address sovereign-bank nexus: - SRB is well-suited in the European context: flexible and can address common exposures unaccounted for in CRR. - Calibrate SRB to increase with the share of domestic sovereign exposures to raise banks’ resilience and incentivize diversification. - Calibration should avoid unintended side-effects by allowing sufficient sovereign debt holdings for liquidity needs and applying gradual capital surcharges. - Complement SRB with measures to diversify bank holdings away from the domestic sovereign: - Expand NBR’s collateral framework to include euro area government bonds as eligible collateral. - Review government debt management strategy. - Enhance the FX swap market. - Apply stressed DSTI limits to all loans and scale back Prima Casa to strengthen LTV limits’ effectiveness. - Enforce currency-differentiated LCR and monitor currency-differentiated NSFR. - Tighten provisioning for NBFLs in line with IFRS9 application to banks. - Require asset management companies to report to the credit registry. ### Key recommendations (Table of FSAP 2018 Recommendations) - 1. Require proposed policy actions and distribution of votes to be publicly disclosed in the summary of meetings. — NBR, MoPF, ASF — NT - 2. Develop a common assessment of systemic risk at each NCMO meeting. — NBR, MoPF, ASF — NT - 3. Apply a stressed DSTI limit to all loans (mortgages and consumer loans). — NBR — NT - 4. Scale back the Prima Casa program. — MoPF — MT - 5. Use the Systemic Risk Buffer (SRB) to address risks from the sovereign-bank nexus. — NBR, MoPF, ASF — NT - 6. Complement the SRB with other measures: expand NBR collateral framework to include euro area government bonds, review debt management strategy, enhance FX swap market. — NBR, MoPF, ASF — MT - 7. Enforce a currency-differentiated LCR for significant currencies. — NBR — NT - 8. Monitor a currency-differentiated NSFR for significant currencies. — NBR — NT - 9. Ensure provisioning requirements for NBFLs tighten with IFRS9 application to banks. — MoPF, NBR — NT - 10. Require asset management companies to report to the credit registry. — NT ### SRB calibration, example schedule, and impact assessment - Rationale and calibration approach: - Apply SRB marginally and only for exposure shares beyond a threshold to preserve holdings needed for liquidity. - Example marginal schedule (Exposure to Sovereign as a Share of RWA (percent) — SRB add-on): - < 15: 0 percent - 15-30: 2 percent - 30-40: 4 percent - 40-50: 6 percent - >50: 8 percent - Worked example: a bank with sovereign exposures of 45 percent of RWA would face a 1.0 percent SRB surcharge (0*15 + 0.02*15 + 0.04*10 + 0.06*5 = 1.0). - Desirable side-effects: - Incentivize intra-financial system trade of sovereign bond holdings to even out exposures and reduce system-wide SRB due to schedule nonlinearity. - Diversify investor base; foreign investors hold approximately 17 percent of domestic sovereign bonds. - Reduce distortion from zero-risk weight on sovereign bonds and support credit growth to the non-financial private sector at the margin. - Impact assessment summary (Annex 1 highlights): - Baseline assumptions: - Increase in required bank capital due to SRB close to 1 percent of RWA or RON 1.7 billion for the largest 12 banks. - Assets of the largest 12 Romanian banks: RON 317 billion as of end–2016. - Exposure to government securities close to 22 percent of total assets, roughly RON 69 billion. - ROE used: 12 percent (end 2017); deposit rates close to 2 percent; tax rate 16 percent. - Baseline quantitative impacts: - Effective increase in CET1 capital close to 1 percent of RWA for the 12 banks (RON 1.7 billion). - Increase in banks’ cost of funding from switching to equity from deposits: RON 174 million annually. - Higher funding costs imply an increase in government yields of just over 30 basis points (bps) — presented as an upper bound. - Reasons 30bps is likely an upper bound: - If cost of equity falls with higher capital (50 percent adjustment), yield increase would be roughly 15bps. - Bank debt may be more expensive than deposits, so deposit-rate assumption is conservative. - Not all additional capital may need to be raised due to existing capital buffers. - Inter-bank redistribution could lower aggregate capital requirements (example: evenly distributed sovereign debt could reduce system SRB to RON 1.2 billion and SRB add-on to 0.7 percent of RWA). - Other investors (parents, pension funds, foreign investors) could increase holdings, muting price impact. - Context: Romania 10-year government bond yields ~4.5 percent during first four months of 2018; a 30bps increase is about 1/16 or less than 7 percent of this level; given relatively low government debt-to-GDP, such increase should not pose material risks to public sector debt sustainability. - Phased-in approach (three years) — example marginal SRB add-on by year: - Year 1 marginal add-ons: <15: 0 percent; 15–30: 1 percent; 30–40: 2 percent; 40–50: 4 percent; >50: 5 percent - Year 2 marginal add-ons: <15: 0 percent; 15–30: 2 percent; 30–40: 3 percent; 40–50: 5 percent; >50: 6 percent - Year 3 marginal add-ons: <15: 0 percent; 15–30: 2 percent; 30–40: 4 percent; 40–50: 6 percent; >50: 8 percent - Estimated additional capital (system-wide): Year 1: 0.6 percent of RWA; Year 2: 0.8 percent of RWA; Year 3: 1 percent of RWA. - Estimated impact on government bond yields: Year 1: ~20 bps; Year 2: 25 bps; Year 3: 30 bps. - Impact on banks’ profitability (ROE) — quantified example: - Assumed ROE: 12 percent; implied ROA for top 12 banks: ~1.3 percent. - Additional capital raised: RON 1.7 billion; debt-to-equity falls from 8.5 to 8.1. - Implied reduced ROE: 11.4 percent (change: 0.6 percentage points lower). - Conclusion: impact on profitability is very limited. - Impact on banks’ liquidity: - Sovereign securities form 78 percent of all lei denominated HQLA. - Under counterfactual where only available HQLA are government securities, most banks would not face a positive capital buffer; SRB buffer for aggregate banking sector ~0.1 percent. - Conclusion: cost associated with meeting LCR imposed by SRB is very small. - SRB surcharge and FX liquidity risks: - Banks have FX denominated deposits of 32 percent of total deposits. - SRB may encourage banks to hold more FX denominated securities, reducing need to extend loans in FX and helping meet HQLA requirements in FX. ### FX liquidity, funding risks, and structural vulnerabilities - FX lending risks: - A revival in FX lending funded by increased reliance on wholesale funding cannot be ruled out. - Almost all banks meet the 100 percent NSFR in RON and euro, but liquidity and funding patterns can change rapidly. - Policy recommendations on FX liquidity and maturity mismatch: - Enforce a currency-differentiated LCR for significant currencies to force holding of HQLA in all significant currencies. - Monitor a currency-differentiated NSFR to assess maturity mismatches in significant currencies. - Implementation options: introduce currency-differentiated LCR as a Pillar 2 measure or as Pillar 1 via Article 458 CRR. - If excessive FX credit growth creates instability, consider enforcing currency-differentiated NSFR and, if insufficient, temporary increase in FX reserve requirements complemented by macroeconomic policy and financial regulations, provided risks stem from a capital inflow surge. - Structural and systemic findings: - Banking system: 37 banks, 29 foreign-owned; banking system holds ~80 percent of financial sector assets; total bank assets ~56 percent of GDP. - Five largest banks account for ~60 percent of total deposits and 57 percent of total loans. - Domestic private sector deposits rose from ~48 percent of banks’ total liabilities in 2011 to ~60 percent in 2016. - Parent funding declined to about EUR 7 billion (about one third of 2011 level). - Interconnectedness: exposures to NBFLs are more important than interbank exposures; some NBFLs have greater systemic impact than banks. - O-SII buffer: nine banks assessed as O-SIIs and subjected to a one percent buffer as of January 1, 2018 (seven foreign- and two domestically-owned); O-SIIs held over 76 percent of bank assets as of March 31, 2017. - Capital conservation buffer phase-in: increments of 0.625 percent each year from January 2016, reaching 2.5 percent on January 1, 2019. - Regulatory vigilance and data gaps: - Authorities should guard against regulatory arbitrage in the NBFL sector; align provisioning regimes with banks. - Require debt collection companies/asset management companies to report to the credit registry to close data gaps from NPLs sold off banks’ balance sheets. *Source: IMF staff analysis and recommendations as presented in the provided content.* ### EXECUTIVE SUMMARY _______________________________________ 5 ### EXECUTIVE SUMMARY ### Summary of systemic risk monitoring and macroprudential capacity - The National Bank of Romania (NBR) has a long experience in implementing macroprudential policy measures and a relatively sophisticated systemic risk monitoring framework, including: - Monitoring several indicators derived from the nation-wide credit register and related data sources. - Constructing summary indicators to facilitate overall risk assessment. - Using various economic models to assess macro-financial developments, shock impacts, and policy actions. - Regular solvency and liquidity stress tests for banks. - Data and information gaps remain, notably because extensive NPL exposures have been sold to asset management companies that do not have to report to the credit registry. ### Institutional framework: mandate, powers, and concerns - Institutional changes: - A new National Committee for Macroprudential Oversight (NCMO) was established in April 2017 and is by law the national macroprudential authority with a clear legal mandate to set macroprudential policies. - The NCMO is chaired by the Governor of the NBR and its Secretariat resides within the NBR. - The nine-member NCMO includes three NBR representatives (reduced from a proposed five), giving the NBR the same number of representatives as the ASF and the Ministry of Public Finance (MoPF). - Functioning and governance: - The NCMO has held five meetings and issued ten recommendations, including required quarterly recommendations on the countercyclical capital buffer. - The Technical Commissions on Systemic Risk and Financial Crisis Management (TSCR and TCFCM) are yet to become operational; their composition was approved by the NCMO in February 2018. - Powers and tools: - The NCMO has direct (hard) powers over a wide range of macroprudential tools, the power to recommend actions to other authorities or Government (with a “comply or explain” mechanism), and the power to issue warnings and opinions. - Key institutional concerns and recommended governance reforms: - There is an underlying policy inaction bias: the cost of policy actions is sooner and more observable than their potential benefits. - NBR can propose policy action and be outvoted without the need for the decision or vote distribution to be disclosed or related to a common overall risk assessment. - Recommended governance changes: - Require proposed policy actions and distribution of votes to be publicly disclosed in the summary of meetings. - Develop a common assessment of systemic risk at each NCMO meeting to foster consensus and common ownership of actions. ### Macroprudential toolkit: current instruments and gaps - Recent framework developments: - The EU Capital Requirements Directive and Regulation (CRD/CRR) framework has become operational in Romania. - Authorities have implemented, or are phasing-in, a number of capital buffers and the Liquidity Coverage Ratio (LCR) under this framework. - Longer-standing borrower tools remain but their scope is narrow. - Existing borrower- and sector-specific calibrations (Table of measures in use): - Maximum LTV on domestic currency-denominated mortgages: "Eighty-five percent for domestic currency-denominated mortgages, except for Prima Casa mortgages." (October 2011) - Maximum LTV on FX mortgages: "Eighty percent for FX loans to hedged borrowers, 75 percent for EUR-denominated mortgages to unhedged borrowers, and 60 percent in other currencies to unhedged borrowers." (October 2011) - Stressed DSTI on consumer loans: "Credit institutions are responsible for testing borrowers’ resilience throughout the duration of the credit, and the maximum indebtedness level should be determined by taking into consideration the interest rate risk, the currency risk, and the risk of a reduction in income." (October 2011) - Other restrictions on consumer loans: "Maturity restricted to five years. Value of goods purchased for consumer credit in FX capped at 133 percent of the credit amount." (October 2011) - Higher risk weights on commercial real estate exposures: "100 percent risk weights on commercial real estate exposures." (January 2014) - Other systemically important institutions (OSII) buffer: "One percent of total risk exposure for nine credit institutions identified as having systemic importance." (January 2018) - Identified toolkit gaps and recommended extensions: - Apply a stressed DSTI limit to all loans (both mortgages and consumer loans) to mitigate risks of excessive credit growth and rising defaults given households’ vulnerability to a rise in interest rates or risk premia. - Gradually scale back the Prima Casa program to mitigate housing sector imbalances and support effectiveness of existing LTV limits. - Enforce a currency-differentiated LCR for significant currencies and monitor a currency-differentiated NSFR. - Strengthen oversight and provisioning for nonbank financial lenders (NBFLs) to reduce scope for regulatory arbitrage, including aligning provisioning regimes with banks. - Require asset management companies (debt collecting agencies specializing in NPL purchases) to report to the credit registry to close data gaps. ### Systemic vulnerabilities and policy implications - Main vulnerabilities identified: - Sovereign-bank nexus: - Romanian banks have exceptionally concentrated and large exposures towards the domestic sovereign, posing important risks for individual banks and the system. - Large and concentrated sovereign exposures can result in sizeable valuation losses if interest rates or sovereign risk premia increase. - Housing and household sector: - Mortgage lending growth and house price developments point to rising vulnerabilities to increases in interest rates. - The Prima Casa program allows borrowing at up to 95 percent LTV for supported mortgages, undermining LTV limits. - Households are vulnerable to interest rate or risk-premia increases, as reflected in mission stress tests. - Corporate sector and funding shifts: - Corporate sector leverage has been increasing despite contracting bank credit extension; corporate external borrowing and reliance on trade credit have increased substantially. - This funding shift could indicate regulatory arbitrage and raises contagion concerns. - FX-related vulnerabilities: - FX loans have subsided but remain relatively large; many corporate FX borrowers are unhedged. - FX liquidity risks can exist within an environment of ample overall liquidity; revival of FX lending reliant on wholesale funding cannot be excluded. - NBFLs and regulatory arbitrage: - Direct external borrowing and lending from NBFLs pose a regulatory arbitrage risk, with rapid lending growth to SMEs and low-income households in the NBFL sector. - Policy implications and recommended measures: - Use the Systemic Risk Buffer (SRB) to address risks from the sovereign-bank nexus: - The SRB is well-suited in the European context: flexible in implementation and suitable to address common exposures unaccounted for in CRR. - Calibrate the SRB to increase with the share of domestic sovereign exposures to raise banks’ resilience and incentivize diversification away from excessive sovereign concentration. - Calibration should avoid unintended side-effects (excessive liquidity reduction, bond market sell-off, or unwarranted macro-financial dynamics) by: - Allowing sufficient sovereign debt holdings to safeguard liquidity without surcharge. - Applying a gradual rise in capital surcharges as sovereign exposures increase while ensuring valuable resilience. - Complement the SRB with measures to diversify bank holdings away from domestic sovereigns, including: - Expansion of the NBR’s collateral framework to include euro area government bonds as eligible collateral. - Further review of government debt management strategy. - Measures to enhance the FX swap market. - Apply stressed DSTI limits to all loans and scale back Prima Casa to strengthen effectiveness of LTV limits and reduce household vulnerabilities. - Enforce currency-differentiated LCR and monitor currency-differentiated NSFR. - Tighten provisioning requirements for NBFLs in line with the application of IFRS9 to banks to prevent regulatory arbitrage. - Require asset management companies to report to the credit registry to improve data quality. ### Key recommendations (from Table 1: FSAP 2018 Recommendations) - 1. Strengthen the NCMO’s accountability, transparency, and coordination frameworks by requiring proposed policy actions and distribution of votes to be publicly disclosed in the summary of meetings. — NBR, MoPF, ASF — NT - 2. Develop a common assessment of systemic risk at each NCMO meeting to foster consensus and common ownership of actions. — NBR, MoPF, ASF — NT - 3. Apply a stressed DSTI limit to all loans (mortgages and consumer loans). — NBR — NT - 4. Scale back the Prima Casa program to mitigate housing sector imbalances and support the effectiveness of the LTV limits. — MoPF — MT - 5. Use the Systemic Risk Buffer (SRB) to address risks stemming from the strong sovereign-bank nexus. — NBR, MoPF, ASF — NT - 6. Complement the SRB with other measures to support diversification away from the domestic sovereign, including expansion of the NBR’s collateral framework to include euro area government bonds, review of government debt management strategy, and measures to enhance the FX swap market. — NBR, MoPF, ASF — MT - 7. Enforce a currency-differentiated LCR for significant currencies. — NBR — NT - 8. Monitor a currency-differentiated NSFR for significant currencies. — NBR — NT - 9. Ensure provisioning requirements for NBFLs tighten in line with the application of IFRS9 to banks to prevent regulatory arbitrage. — MoPF, NBR — NT - 10. Require asset management companies to report to the credit registry to improve data quality. — NT *International Monetary Fund — Technical Note prepared by Thorvardur Tjoervi Olafsson, Monetary and Capital Markets Department.* ### 1.875 percent of total exposure and set to gradually rise ### 1.875 percent of total exposure and set to gradually rise ### C. Principle 3: Effective Coordination and Cooperation - The functioning of the NCMO is still being established: - As of April 2018, it has held five meetings and issued ten recommendations, including quarterly recommendations on the countercyclical capital buffer (CCyB). - Representatives of the MoPF in the NCMO have changed frequently due to changes in government. - The Technical Commissions on Systemic Risk and Financial Crisis Management (TCSR and TCFCM) are still to become operational. - The accountability framework (publication of a policy strategy, summary of meetings, issued recommendations and warnings, and an annual report) is still being developed. - The NCMO does not reach a common assessment of systemic risks at its meetings: - The NCMO has relatively strong powers to gather data and few legal impediments for sharing information between the NBR and the ASF. - The TCSR is not assumed to prepare an assessment of systemic risk ahead of NCMO meetings. - The NCMO does not strive to reach a consensus on key risks at its meetings, risking oversight of vulnerabilities, weakening common ownership of policy decisions, and leaving the opportunity to communicate a clear view of systemic risks unutilized. ### D. Recommendations (institutional and governance) - Strengthen NCMO accountability, transparency, and coordination frameworks to counteract inaction bias: - Require proposed policy actions and distribution of votes to be publicly disclosed in the summary of meetings. - Allow for exemptions to public disclosure where unanimous NCMO vote deems disclosure likely detrimental to financial stability. - Develop a common assessment of systemic risk at each NCMO meeting: - The TCSR, chaired by the NBR, should prepare a draft assessment of key risks for NCMO discussion and agreement. - Publish the NCMO’s assessment of systemic risk in the summary of meetings to create a clear relation between assessment and policy actions and to foster public and market understanding. ### Systemic Risk Monitoring - Macroprudential policy requires guided discretion: combine key indicators with judgement, data, qualitative information, and analytical capacity to map risk assessment into policy action. - NBR capacities and outputs: - The NBR has long experience in implementing macroprudential measures and a relatively sophisticated systemic risk monitoring framework. - Monitors several indicators derived from the nation-wide credit register and related data sources; constructs summary indicators. - The biannual Financial Stability Report is the main communication tool prepared by the Financial Stability Department. - Uses economic models to assess macro-financial developments, shocks, and policy effects; conducts solvency and liquidity stress tests. - ASF capacities: - The ASF monitors financial markets, insurance sector, pension funds, and asset managers. - Founded in 2013 by merging three sectoral agencies; has less experience but is enhancing its capacity and publishes regular sector reports with a dedicated risk assessment unit. - TCSR potential and composition: - TCSR could ensure comprehensive systemic risk assessment based on NBR and ASF work and improve coordination. - Composition decided at NCMO meeting in February 2018: seven members; chaired by Director of the Financial Stability Department at the NBR; NBR has three members; ASF and MoPF two members each. - Data quality and gaps: - Credit register provides detailed loan-level information supporting monitoring and research. - House price information has improved but series are still relatively short. - Data on commercial real estate prices, transactions volumes, and exposures is scarce. - A key data gap arises from NPL exposures sold to foreign-owned asset management companies, which do not have to report to the credit registry. - Recommendation on data reporting: - Require asset management companies to report to the credit registry to ensure accurate information on revised principal and payment profiles, and to avoid misreporting of total indebtedness and debt service burden; achieveable by amending primary legislation. ### SYSTEMIC RISKS AND MACROPRUDENTIAL TOOLS - Objective: - Map systemic vulnerabilities into recommendations for the macroprudential policy toolkit using multiple signaling indicators and FSAP risk analysis. A. Broad-Based Vulnerabilities - Credit developments: - Credit-to-GDP gap has been in negative territory for several years. - Gap for shorter credit cycles has been gradually closing but remains well below the lower 2 percent threshold indicating need to start increasing buffers if supported by other surveillance information. - Credit growth recently picking up after prolonged contraction, especially due to increased mortgage lending. - Funding now primarily from local deposits rather than parent and wholesale funding. - Capital ratios are higher, making the banking system more resilient to losses. - CCyB calibration: - CCyB rate has been kept at zero percent. - NCMO assesses CCyB quarterly and has recommended the NBR monitor household indebtedness. - Asset quality: - NPLs peaked at 21.9 percent of total loans in 2013. - An action plan removed roughly €4 billion in NPLs from banks’ balance sheets. - In December 2017 NPLs had declined to 6.4 percent of total loans. - Provisioning ratio (including general provisions) is high at around 65 percent. - Recommendation: keep existing supervisory measures, including bank-specific reduction targets, until NPL share reaches an appropriately low level. B. Vulnerabilities from Housing and Household Sector - Mortgage lending and Prima Casa program: - Prima Casa: State guarantees 50 percent of the mortgage; down payment 5 percent of property value; interest rate capped at ROBOR plus 2.5 percent. - Program has been in place and renewed annually since 2009 and accounts for the majority of mortgages extended by banks. - Housing loans increased from 21 percent of loans to households to more than 58 percent between 2008 and 2016. - House prices and mortgage characteristics: - Annual house price growth rate of 6.2 percent as of September 2017. - Prices still well below historical peaks; a house price correction seems unlikely. - Large majority of mortgage contracts are at variable rates, raising vulnerability to interest rate increases. - Stress-test findings: - In a scenario with rising interest rates and a sharp output contraction, real estate mortgage losses account for a majority of overall credit losses in the solvency stress test; followed by losses from loans to small-to-medium-sized enterprises backed by real estate. - Macroprudential toolkit recommendations: - Apply clear maximum DSTI limits to mortgages, including Prima Casa loans: - In September 2017, about 25 percent of debtors had a DSTI level in excess of 55 percent. - A stressed maximum DSTI ratio is currently only applied to consumer loans. - Calibration could draw on joint analysis using loan-level information from the credit register. - Gradually scale back the Prima Casa program: - Prima Casa allows LTV ratios up to 95 percent, undermining effectiveness of existing macroprudential tools. - Authorities started to scale back the program; a strategic review is planned in 2019. - Continued unwinding is important to mitigate housing sector imbalances and support effectiveness of LTV limits. C. Vulnerabilities from Sovereign Exposures - Sovereign exposure levels and concentration: - As of December 2016, sovereign debt exposure of banks was around 20 percent of assets, up from below 5 percent in 2008. - Exposure increase has been rapid and steady over seven years. - Duration of domestic sovereign debt held by banks (primarily in trading book) is long relative to banks’ funding, exposing banks to interest rate risk. - Government guarantees under Prima Casa further reinforce indirect banking sector exposure to the sovereign. - Sovereign exposures are exceptionally concentrated towards the domestic sovereign compared with other EU countries. - Stress-test results and risks: - FSAP risk analysis: in the adverse scenario, a total drop of 300bps in capital over the 3-year horizon due to sovereign exposures. - Trading gains of 0.6 percent of RWA in 2017 turn to a market loss of 3.6 percent in 2018 for the system as a whole (see Technical Note on the Risk Analysis). - Impacts can be larger and comparable in magnitude to credit losses under a severe economic recession for some banks. - Sovereign-bank nexus despite low debt: - Public debt roughly 36.8 percent of GDP at year-end 2017. - NBR analysis points to debt levels between 40 and 45 percent of GDP as thresholds for a material increase in the likelihood of economic recession and detrimental impact on debt refinancing capacity. - Procyclicality of fiscal policy can lead to tighter monetary policy and higher interest rates, increasing risk premia and volatility that challenge financial stability. - Policy recommendations to mitigate sovereign-bank nexus: - Introduce prudential policy measures to increase loss-absorbing capacity and disincentivize excessive concentration of sovereign exposures while avoiding unintended side-effects (excessive reduction of liquidity, bond market sell-offs, or other unwarranted macro-financial dynamics). - Use the Systemic Risk Buffer (SRB) to address the sovereign-bank nexus: - SRB is flexible in implementation and suitable to address common exposures and risks unaccounted for in the CRR framework. - SRB can increase banks’ resilience to sovereign losses and encourage limits on excessive concentration of sovereign exposures. - SRB can be relaxed to mitigate risks of detrimental procyclical dynamics. - Costs can be limited by careful and appropriately gradual calibration. *Source: IMF staff analysis and recommendations as presented in the provided content.* ### 35. A gradual and marginal SRB calibration is appropriate to avoid unintended side- ### 35. A gradual and marginal SRB calibration is appropriate to avoid unintended side-effects ### Example calibration of the SRB - Calibration applies a positive SRB only for exposure shares beyond a threshold, recognizing some domestic government bonds are held to fulfill liquidity requirements. - Marginal and gradual rise in capital surcharges as sovereign exposures increase relative to RWA, reaching around 3 percent surcharge for banks with high exposure shares (text: "sufficient surcharge (of around 3 percent) to ensure valuable resilience"). - Example marginal schedule (Exposure to Sovereign as a Share of RWA (percent) — SRB add-on): - < 15: 0 percent - 15-30: 2 percent - 30-40: 4 percent - 40-50: 6 percent - >50: 8 percent - Worked example: a bank with sovereign exposures of 45 percent of RWA would face a 1 percent SRB surcharge (calculation shown in source: 0*15+0.02*15+0.04*10+0.06*5=1.0). ### Desirable side-effects of a gradual/marginal SRB calibration - Incentive for intra-financial system trade of sovereign bond holdings: - Banks with high exposure face higher buffer rates and therefore have incentives to trade with banks with smaller exposures and lower buffer rates. - More even distribution of sovereign holdings would lower system-wide capital surcharges due to nonlinearity in the schedule. - Intra-financial system trade would limit potential for a broad-based bond market sell-off, as such trade can effectively reduce the SRB. - Diversification of investor base if banks sell holdings or reduce share of new issues purchased: - Investor base could include a higher share of other domestic institutional investors as well as foreign investors. - Source notes foreign investors hold approximately 17 percent of domestic sovereign bonds. - Such investors often have bigger appetite for long-durations, allowing government to lengthen debt maturity and reduce rollover risks. - Reduction of distortion from zero-risk weight on sovereign bonds: - SRB would mitigate incentive to expand sovereign portfolios at expense of lending to the real economy and could at the margin support credit growth to the non-financial private sector. ### Impact assessment summary - Annex 1 analysis: limited effects expected on sovereign bond yields, banks’ profitability, and liquidity. - Extreme case where banks fully pass increased funding cost to sovereign: - Bond yields likely to increase by only up to around 30 basis points (bps). - This is presented as an upper bound; mitigating factors likely reduce the increase. - Opposite extreme where banks fully absorb increased funding costs: - Impact on banks’ profitability would be very limited. - Effects on banks’ liquidity from introducing the SRB are found to be small. - Annex also provides an example of gradual phasing-in of SRB over a period of three years. ### Notification, thresholds, and EU procedures - Before setting or adjusting the SRB, national authorities must notify: - European Commission, European Systemic Risk Board (ESRB), European Banking Authority (EBA), and authorities of other Member States concerned, with at least one-month notice. - Notification must justify why other CRD/CRR measures are insufficient to address identified systemic risk. - Notification sufficient up to SRB level of 3 percent on entities which are subsidiaries of parent banks from other Member States; up to 5 percent on entities which are not subsidiaries of parent banks from other Member States. - European Commission approval is needed for SRB levels above these thresholds. ### Complementary and alternative approaches (summary of Box 1) - Higher Risk Weights: - Legally constrained in CRD/CRR; exposures to Member States’ central governments and central banks in domestic currency are assigned a zero RW (Article 114 of the CRR). - CRR lists limited avenues to implement higher RWs (Article 124(2), Article 458, Article 103). - Large Exposure Limits (LELs): - General LEL at 25 percent of eligible capital (or EUR 150 million, whichever higher); zero-RW sovereign exposures are exempt. - Legal restrictions limit use; exemption on zero-RW sovereign holdings cannot be removed. - Liquidity Requirements: - Haircuts on domestic sovereign bond holdings in liquidity buffers face legal and effectiveness challenges; Level 1 status for zero-RW government bonds in Liquidity Coverage Ratio regulation prevents haircuts. - Would not ensure additional capital to absorb losses during sovereign distress. - Pillar 2 Measures: - Flexible and can target individual institutions or groups via SREP or supervisory stress tests. - Drawbacks: lack of transparency and nondisclosure; implementation risks and potential for less orderly reallocation compared to transparent SRB calibration. ### Policy recommendations and complementarities - Preferably complement SRB with other measures: - Expand NBR’s collateral framework to include euro area government bonds (potentially with minimum credit ratings) as eligible collateral to support diversification away from domestic sovereign. - Further reforms to government debt management strategy to limit impact of SRB on government bond yields. - Enhance FX swap market to help banks manage potentially larger FX risks when reallocating portfolios towards foreign bonds; would also help limit currency mismatch risks in non-financial private sector. - NCMO context: - In its December 2017 meeting, the NCMO recommended use of the SRB to facilitate continued reduction of NPLs; while NPLs fell below eight percent, high concentration of sovereign exposures remained largely unattended. - Potential option: utilize SRB to ensure further progress in lowering NPLs before applying SRB to mitigate concentrated sovereign exposures. ### Related financial sector vulnerabilities noted in the chapter - Corporate sector: - Bank exposure to corporate sector has decreased while corporate leverage remained high and increased recently. - Corporate reliance on external borrowing and domestic funding through trade credit increased; corporate leverage increased between 2012 and 2016 despite decline in bank credit. - More than 40 percent of firms have negative equity; about 30 percent of these companies were established over the last five years. - NCMO set up a working group to assess corporate financial position and measures to address high share of technically insolvent firms. - FX-related vulnerabilities: - FX loans contracted but remain material: FX loans were 44 percent and 42 percent of total corporate and household loans, respectively, in 2017. - NPLs higher for FX loans than domestic currency: 12.3 percent vs 3.3 percent in December 2016. - Non-bank financial lenders (NBFLs) lending to corporates predominantly in FX (84 percent). - Household borrower limits (LTV and DSTI) are tighter for FX lending; DSTI uses a stressed scenario including sizable depreciation. - NBR should enhance supervision of NBFL sector. - Funding and liquidity: - Aggregate banking sector liquidity abundant, partly due to sizable sovereign exposures. - Funding structure similar in RON and EUR with retail deposits largest funding source; secured funding almost non-existent. - Currency-differentiated liquidity pockets: some banks meet LCR in RON or EUR but not both; some assessed banks fall short of 100 percent LCR in euro, and a few in RON. *Source: IMF staff calculations and chapter text.* ### 47. A revival in FX lending funded by increased reliance on wholesale funding cannot be ### 47. A revival in FX lending funded by increased reliance on wholesale funding cannot be ### FX liquidity and funding risks — findings - A revival in FX lending funded by increased reliance on wholesale funding cannot be ruled out. - Banks’ funding patterns have become more stable: almost all the banks meet the 100 percent net stable funding ratio (NSFR), both in RON and euro. - Experience shows liquidity and funding patterns can change rapidly and need close monitoring and mitigation if needed. ### Policy recommendations on FX liquidity and maturity mismatch - Enforce a currency-differentiated LCR for significant currencies to strengthen banks’ resilience towards liquidity shocks by forcing them to hold high quality liquid assets (HQLA) in all significant currencies. - Monitor a currency-differentiated NSFR to assess maturity mismatches in all significant currencies (e.g., due to overreliance on wholesale funding). - Implementation options: - Introduce a currency-differentiated LCR as a Pillar 2 measure, or as a Pillar 1 measure following the process in Article 458 CRR. - If excessive FX credit growth creates financial instability risks: - Enforce the currency-differentiated NSFR; and if insufficient, consider a temporary increase in FX reserve requirements to complement appropriate macroeconomic policy adjustment and financial regulations, provided risks stem from a capital inflow surge. ### Structural vulnerabilities — key findings - Concentration and market structure: - The banking system consists of 37 banks of which 29 are foreign-owned. - The banking system holds around 80 percent of financial sector assets. - Total bank assets amount to about 56 percent of GDP. - The five largest banks account for about 60 percent of total deposits and 57 percent of total loans. - Deposit and parent funding trends: - Domestic private sector deposits increased from about 48 percent of banks’ total liabilities in 2011 to about 60 percent in 2016. - Parent funding declined to about EUR 7 billion (about one third of the level in 2011). - Interconnectedness: - Banks’ exposures to NBFLs are more important than interbank exposures; some NBFLs have a greater systemic impact on the network than banks. - Parent banks are particularly vulnerable to distress from NBFLs they own due to potential reputational effects and depositor runs. - Systemic importance and buffers: - Nine banks are assessed to be systematically important and are subjected to a one percent O-SII-buffer. - From January 1, 2018 seven foreign- and two domestically-owned credit institutions are assessed to be O-SIIs. - O-SIIs held over 76 percent of bank assets and similar shares of non-financial private sector loans and deposits as of March 31, 2017. - O-SIIs’ share in banks’ cross-border assets and liabilities was 84 and 77 percent, respectively. - Intra-financial assets and liabilities shares for O-SIIs were 59 and 75 percent, respectively. - Capital conservation buffer: - Phase-in began in January 2016 with equal increments of 0.625 percent each year, resulting in the buffer reaching the uniform 2.5 percent level on January 1, 2019. ### Regulatory vigilance and data gaps - Authorities need to exhibit vigilance towards regulatory arbitrage, especially in the NBFL sector where lending to SMEs and low-income households was increasing and mainly funded by banks. - The NBR introduced measures to strengthen oversight over the NBFL sector and should align the sector’s provisioning regime with that of banks. - Data and information gaps remain (e.g., sold household and corporate NPLs are off banks’ balance sheets); authorities should require debt collection companies to report to the credit registry to improve data quality. ### Conclusion — macroprudential institutional framework and toolkit - The NCMO was established in April 2017 as the national macroprudential authority with a clear legal mandate to set macroprudential policies to safeguard financial stability. - The NCMO has direct powers over a wide range of macroprudential tools, can recommend actions to other authorities, issue warnings with a ‘comply or explain’ mechanism, and is subject to an accountability framework (publication of policy strategy, summary of meetings, issued recommendations and warnings, and an annual report). - The NCMO had held five meetings and issued 10 recommendations, including on the CCyB. - The NBR has extensive experience in macroprudential policy implementation and systemic risk monitoring, using indicators derived from the nationwide credit registry and regular stress tests. ### Macroprudential toolkit recommendations - Apply a stressed DSTI limit to all loans (mortgages and consumer loans) to mitigate risks of excessive credit growth and rise in defaults given household vulnerabilities to interest rate or risk premia increases. - Calibrate DSTI using loan-level information from the Romanian credit register (joint analysis with the NBR). - Gradually scale back the Prima Casa program to mitigate housing sector imbalances and support the effectiveness of LTV limits. - Enforce a currency-differentiated LCR for significant currencies and monitor a currency-differentiated NSFR to address FX liquidity risks that can exist despite ample overall liquidity. - Calibrate the SRB carefully to address risks from the strong sovereign-bank nexus while mitigating potential unintended side-effects. ### Annex I — Impact of the SRB scheme on bond yields and bank profitability: key figures and scenarios - Methodology and baseline assumptions: - Assumes increase in required bank capital due to the SRB is close to 1 percent of RWA or RON 1.7 billion for the largest 12 banks. - Assets of the largest 12 Romanian banks stood at RON 317 billion as of end–2016. - Exposure to government securities was close to 22 percent of total assets, or roughly RON 69 billion. - Assumes increase in funding costs is fully passed into bond prices and there is no adjustment in quantity of bonds held. - Uses ROE of 12 percent (end 2017) and deposit rates close to 2 percent on average; tax rate considered is 16 percent. - Baseline quantitative impacts: - Calibrated SRB implies an effective increase in CET1 capital of close to 1 percent of RWA for the 12 banks, or RON 1.7 billion. - Increase in banks’ cost of funding from switching to equity from deposits amounts to RON 174 million annually. - Higher funding costs imply an increase in government yields of just over 30bps (assumed uniform across maturities). - Reasons the 30bps estimate is likely an upper bound: - If cost of equity falls with higher capital (a 50 percent adjustment between Modigliani-Miller and baseline), increase in yields would be roughly 15bps. - Bank debt may be more expensive than deposits, so using deposit rates is conservative and actual marginal funding cost increase could be smaller. - Not all additional capital may need to be raised because many banks have significant capital buffers above regulatory requirements. - Inter-bank trade and redistribution of sovereign holdings could lower aggregate capital requirements: if sovereign debt were evenly distributed, overall capital requirements due to SRB could decline to RON 1.2 billion and SRB rate add-on to 0.7 percent of RWA (compared to RON 1.7 billion and 1.0 percent of RWA in the baseline). - Other investors (parent banks, pension funds, investors abroad not subject to SRB) could increase holdings given higher yields, muting price impact. - Context and debt sustainability: - Romania 10-year government bond yields fluctuated around 4.5 percent during the first four months of 2018; a 30bps increase is about 1/16 or less than 7 percent of this level. - Given relatively low government debt-to-GDP, the analysis suggests such an increase should not pose material risks to public sector debt sustainability. - Consistency with literature: - Literature estimates limited impacts of higher capital requirements on lending costs (examples: Basel Committee (2010) 9–20bps for a 1 percent increase; Schanz et al. (2011) find 7.4bps for UK; Elliott (2009) and Kashyap, Stein, and Hanson (2010) find limited impacts for the US). *Source: NBR and IMF (chapter content).* ### 10. In this exercise, the impact on banks’ profitability from introducing the proposed SRB ### 10. In this exercise, the impact on banks’ profitability from introducing the proposed SRB ### Impact on profitability - Profitability measure: ROE. - Opposite extremes compared to previous exercise: - Previous extreme: banks pass on the entire increase in the cost of funding to the sovereign, ROE remains constant. - Current extreme: banks absorb the entire increase in funding costs, reflected in a lower ROE. - Assumption: bank return on assets (ROA) remains the same. - Key identity used (as presented): - 푅푅퐶퐶 푅푅∗�1 + 푓푓푑푑푑푑퐶퐶 푑푑퐸퐸푓푓푓푓퐶퐶퐸퐸� = 푅푅퐶퐶퐸퐸 - Assumption: banks do not change asset structure as a result of the SRB capital surcharge; therefore ROA for banks remains the same. - Quantified impact: - Assumed ROE: 12 percent. - Current leverage ratio of the banking sector implies ROA for the top 12 banks is approximately 1.3 percent. - Additional capital raised due to the SRB requirement: RON 1.7 billion. - Debt to equity ratio falls from 8.5 to 8.1. - Implied reduced ROE: 11.4 percent. - Change in ROE: 0.6 percentage points lower compared to the current level. - Conclusion: the impact of introducing the proposed SRB on bank profitability is very limited. ### A Phased-in Approach - Rationale: limit initial impact on government bonds and account for ex ante uncertainty of actual impacts. - Suggested phase-in period: three years. - Example phased-in calibration (from Appendix Figure 1) — marginal SRB add-on by exposure to sovereign as a share of RWA (percent): - Exposure buckets: < 15; 15–30; 30–40; 40–50; >50 - Year 1 marginal SRB add-on: 0 percent; 1 percent; 2 percent; 4 percent; 5 percent - Year 2 marginal SRB add-on: 0 percent; 2 percent; 3 percent; 5 percent; 6 percent - Year 3 marginal SRB add-on: 0 percent; 2 percent; 4 percent; 6 percent; 8 percent - Estimated additional capital (system-wide) under the phase-in: - Year 1: 0.6 percent of RWA - Year 2: 0.8 percent of RWA - Year 3: 1 percent of RWA (ultimate level presented in Appendix Figure 1) - Estimated impact on government bond yields under the phase-in: - Year 1: on the order of 20 bps - Year 2: 25 bps - Year 3: 30 bps ### The Impact on Banks’ Liquidity Positions - Context: sovereign securities form 78 percent of all lei denominated HQLA. - Assessment approach: - Calculate for each bank the amount of HQLA needed to reach a 100 percent LCR on a lei basis. - Counterfactual assumption: the only available HQLA is government securities. - Compute the capital surcharge for each bank associated with holding that amount. - Findings: - Under these assumptions, most banks would not face a positive capital buffer. - The SRB buffer for the aggregate banking sector would amount to just 0.1 percent. - Conclusion: the cost associated with meeting the LCR imposed by the SRB is very small. ### SRB surcharge and FX liquidity risks - Observation: banks in Romania have significant FX denominated deposits: 32 percent of total deposits. - Implications: - To hedge currency risk associated with these liabilities, a significant part of loans—particularly mortgages—are also issued in FX, exposing banks to credit risk associated with unhedged borrowers. - The SRB surcharge could, at the margin, encourage banks to hold more FX denominated securities, which would: - Reduce the need to extend loans in FX given a certain level of FX denominated savings in the economy, as banks can use foreign bonds to close their net open position. - Help banks meet HQLA requirements in FX by holding highly rated foreign bonds, aligning with the team’s liquidity recommendation. *Source: IMF staff calculations and analysis as presented in the chapter.* --- _Source: https://www.imf.org/-/media/files/publications/cr/2018/cr18164.pdf_