Financial Stability Report 2015/2016

Financial Stability Report 2015/2016 Press Conference Presentation Miroslav Singer Governor Prague, 14 June 2016

Structure of presentation I. Overall assessment of risks and setting of countercyclical capital buffer II. Developments and recommendations regarding risks associated with loans secured by property III. Stress testing of bank solvency, bank liquidity, pension management companies and public finances

I. Overall assessment of risks and setting of countercyclical capital buffer

Aggregate macroprudential dashboard (I) Backward-looking perspective (where we have moved) A further strengthening of the resilience of the Czech financial sector to potential adverse shocks: a rise in capital adequacy, favourable liquidity developments, positive changes in risk management by financial institutions. Financial sector resiliance Cyclical risks Structural risks 4

Aggregate macroprudential dashboard (II) Financial sector resiliance Cyclical risks Structural risks Forward-looking perspective (how strong are the potential sources of risk) An increase in cyclical risks: a recovery in demand for loans and an easing of credit standards, extraordinarily low interest rates on loans and a decline in interest margins, growth in property prices and a further drop in returns on high-quality assets. A very moderate rise in structural risks: an increase in maturity transformation of bank deposits and loans, a rise in the share of non-residents on the domestic government bond market. 5

Assessment of position in financial cycle The financial cycle in the Czech Republic is in a growth phase: The aggregate indicator of the financial cycle is rising gradually. Banks lending standards continue to ease. FCI, aggregate indicator of financial cycle (0 minimum, 1 maximum) 0,5 General lending standards in the Czech Republic (difference in market share of banks in pp) Nízké 40 0,4 [1] 20 Část 0 0,3-20 0,2-40 -60 0,1 0,0 06/03 12/05 06/08 12/10 06/13 12/15 Source: CNB -80 09/12 03/13 09/13 03/14 09/14 03/15 09/15 03/16 Loans for house purchase Consumer and other loans Sole proprietors Non-financial corporations Source: Bank Lending Survey, CNB 6

Growth in bank loans and property prices New loans of households for housing and consumption are growing fast. Apartment prices are starting to approach historical highs. Rates of growth of new bank loans for households (year-on-year, %) 50 40 30 Apartment prices transaction and asking prices (cyclical maximum = 100) 110 100 90 20 80 10 70 0-10 -20 03/12 03/13 03/14 03/15 03/16 New loans for house purchase New consumer credit Source: CNB 60 50 03/06 03/08 03/10 03/12 03/14 03/16 Asking prices Prague Asking prices rest of CZ Transaction prices Prague Transaction prices rest of CZ Source: CZSO, CNB calculation 7

Setting of countercyclical capital buffer II/2016 The CNB Bank Board decided to leave the countercyclical capital buffer rate at 0.5% for the time being. There has been no significant change in cyclical risks implying a sizeable increase in systemic risk since the last decision on the setting of the buffer rate was made. If credit growth remains high, credit standards ease further and investor optimism continues to grow, the CNB will stand ready to increase the buffer rate further. Current and announced CCyB rate in the Czech Republic (% of total risk exposure) 1,50 1,25 1,00 0,75 0,50 0,25 0,00-0,25-0,50 10/15 01/16 04/16 07/16 10/16 01/17 04/17 07/17 Source: CNB Current CCyB rate Planned CCyB rate 8

II. Developments and recommendations regarding risks associated with loans secured by property

Developments on mortgage credit market Interest rates on new loans for house purchase decreased to the level usual for loans to non-financial corporations. Institutions provided a larger number of loans with higher individual loan amounts amid declining rates. The annual growth rate of genuinely new mortgage loans was around 30% in 2015 and around 20% in 2016 Q1. Conditions are being created for the emergence of a spiral between property prices and property purchase loans. Retail rates on new CZK loans (%) 5,0 4,5 4,0 3,5 3,0 2,5 2,0 03/11 03/12 03/13 03/14 03/15 03/16 New loans for house purchase New loans to non-financial corporations Source: CNB 10

Growth in household debt The growth rate of loans to households continued to exceed the growth rate of their disposable income. The difference between the growth rates of the two variables increased in 2015 H2. Household debt in the Czech Republic increased to around 60% of GDI (euro area average: around 100% of GDI). The net interest burden on households is also rising. This makes households more sensitive to potential income and interest rate shocks. Rate of growth of loans and nominal GDI of households (year-on-year, %) 19 15 11 7 3-1 12/08 12/09 12/10 12/11 12/12 12/13 12/14 12/15 Rate of growth of loans, total Rate of growth of nominal GDI Source: CNB Note: Both bank and non-bank loans are included. 11

The CNB s 2015 Recommendation Last year, the CNB issued a Recommendation on the management of risks associated with the provision of retail loans secured by residential property. The Recommendation set quantitative LTV limits and qualitative criteria for prudential provision of such loans. The CNB has undertaken to regularly assess compliance with the Recommendation and to tighten the parameters of individual recommendations if increased risks are identified. The CNB conducted a detailed survey of the credit characteristics of individual loans secured by residential property provided newly in 2015 H2 (totalling CZK 117 billion): LTV (loan-to-value), LTI (loan-to-income), DSTI (debt service-toincome), applicant s income. 12

Overall assessment of compliance with Recommendation Positive conclusions of the assessment: Institutions are largely compliant with the recommendations. Institutions are compliant with the 10% limit on the volume of new loans with an LTV of 90% 100% at an aggregate level. Risks identified: However, some institutions are providing some new loans with LTVs of over 100% and exceeding the 10% limit on new loans with an LTV of 90% 100%. Part of new loans have riskier characteristics relating to the client s ability to repay the loan in an unfavourable economic situation. In the category with an LTV of 80% 90%, almost a third had both high LTI and high DSTI levels, of which almost a third were provided to applicants with net monthly income of less than or equal to CZK 25,000. 13

Assessment of compliance with recommendations A, B and C 4% of all new loans were provided with an LTV of over 100%. 9% of new loans were provided with an LTV of 90% 100% at an aggregate level (the quarterly limit is 10%). The 10% quarterly limit on loans with an LTV of 90% 100% was exceeded for 3% of all loans granted in the given period at institution level. The share of new loans with high LTI ratios (over 5) was 45%. Loans with high DSTI levels (above 40%) accounted for 13%; almost all of them had a high LTI. For one-third of loans with both characteristics, applicants had net income of less than or equal to CZK 25,000 Although the average loan repayment period was 24 years, there are institutions in respect of which a significant proportion of loans have been provided for more than 30 years. 14

Assessment for other recommendations The assumption that refinanced loans with an increase (6% of the total) and buy-to-let loans (4% of the total) have riskier characteristics is confirmed. They often have higher LTV, LTI and DSTI levels. LTV distribution of new loans from various perspectives (x-axis: LTV in %; y-axis: share of loans in %) 40 35 30 25 20 15 10 5 0 < 50 50 60 60 70 70 80 80 90 90 100 > 100 Total Refinanced with increase Through intermediaries Buy-to-let Source: CNB DSTI distribution of new loans from various perspectives (x-axis: DSTI in %; y-axis: share of loans in %) 40 35 30 25 20 15 10 5 0 Source: CNB < 20 20 30 30 40 40 50 50 60 > 60 Total Refinanced with increase Through intermediaries Buy-to-let 15

Overall assessment of developments in area of Recommendation The CNB does not assess the aforementioned risks as an acute market overheating giving rise to a direct risk to financial stability.... however, credit standards are highly relaxed and some institutions may be taking on increased risks. The perceived profitability of buying residential property on credit may partly be an illusion. Low interest rates and softer credit standards are fostering growth in property prices, which is then reflected in the size of the loans needed to finance the property purchase and hence a rise in households debt-to-income ratio. At the same time, the growth in residential property prices may not be fully grounded in fundamentals. The Recommendation must be modified mainly by lowering the maximum LTV levels. 16

Changes in Recommendation (I) Recommendation A compliance with LTV limits for new loans: The upper LTV limit of 100% will be reduced to 95% on 1 October 2016 and to 90% on 1 April 2017. The limit of 10% of new loans with an LTV of 90% 100% will change to a limit of 10% of new loans with an LTV of 85% 95% on 1 October 2016. The limit will be set at 15% of new loans with an LTV of 80% 90% on 1 April 2017. The CNB will apply its supervisory powers in respect of institutions exceeding the recommended limits, in particular by setting additional Pillar 2 capital requirements. 17

Changes in Recommendation (II) Recommendation F financing of buy-to-let purchases of residential property: The amount of buy-to-let loans reported by institutions is low, but indications from the market suggest that the volume of such transactions is considerably higher. Institutions are hence recommended to use all available information to determine whether a loan is being used to finance owner-occupied housing or as an investment. If an investment type of loan shows a combination of characteristics with a higher risk level, they should apply an LTV of 60% at most. 18

Other plans in area of Recommendation (I) The CNB will analyse loans provided by building societies in detail from the point of view of their economic nature. The CNB sees increased risks to the debt servicing ability of clients with low income. It will prepare for the introduction of quantitative LTI and DSTI recommendations in 2018 at the latest. The CNB will seek enactment of the power to set LTV, LTI and DSTI limits and, where appropriate, other risk parameters for house purchase loans. Such powers should enable the CNB to respond more effectively to emerging risks to financial institutions and consumers which might arise from excessive softening of credit standards. The alternative would be strong measures in the area of institutions capital. 19

Commercial property market The CNB conducted surveys of new loans secured by commercial property (including residential development projects) since 2014 H2. The commercial property market is also showing signs of increased activity and a fast-growing volume of loans. The highest year-on-year growth was recorded for loans for investment in office property. Amount of new loans secured by commercial property (CZK billions) 50 45 40 35 30 25 20 15 10 5 0 2014 H2 2015 H1 2015 H2 Office investment Office construction Industrial investment Industrial construction Retail investment Retail construction Residential construction Source: CNB 20

Risk characteristics of loans for financing commercial property Loans for financing office property were identified as having potentially the highest risks. Loans from the survey with an LTV of over 70% and, at the same time, a relatively low debt-service coverage ratio (DSCR) of up to 1.2 accounted for 35% of new loans. The property used as collateral is usually the only source of income to service debt. Offices have a relatively high vacancy rate. Share of loans with an LTV of more than 70% and a DSCR of less than or equal to 1.2 broken down by collateral segment (%) 40 35 30 25 20 15 10 5 0 Source: CNB Office Industrial Retail Residential 21

III. Stress testing of bank solvency, bank liquidity, pension management companies and public finances

Baseline Scenario: Alternative scenarios based on the CNB s May forecast published in IR II/2016, assumes growth of the domestic economy in 2016 2018 and a decrease in unemployment at the scenario horizon. Adverse Scenario: a return of a V-shaped recession with a marked decline in GDP, a temporary fall of the economy into deep deflation, a sharp rise in government bond yields, a sizeable increase in unemployment and a drop in wages leads to the exhaustion of the financial reserves of some households and corporations, the negative developments cause a pick-up in credit risk and high credit losses. 23

Macroeconomic developments, credit risk and stress scenario (I) The Adverse Scenario captures above all the risk of a deterioration in loan portfolio quality stemming from: a fall in economic activity, a sizeable decrease in nominal income. Alternative scenarios: real GDP growth (year-on-year change in %) 5 3 1-1 -3-5 -7 03/13 03/14 03/15 03/16 03/17 03/18 03/19 Alternative scenarios: inflation (year-on-year change in %) 5 4 3 2 1 0-1 -2-3 -4-5 inflation target -6 03/13 03/14 03/15 03/16 03/17 03/18 03/19 Baseline Scenario Adverse Scenario Baseline Scenario Adverse Scenario Source: CNB Source: CNB 24

Macroeconomic developments, credit risk and stress scenario (II) The Adverse Scenario involves a strong shock to unemployment and property prices. Alternative scenarios: unemployment (%) 11 Alternative scenarios: property prices (index; 2007Q4 = 100) 130 10 9 8 7 6 5 4 120 110 100 90 80 70 3 03/13 03/14 03/15 03/16 03/17 03/18 03/19 60 03/13 03/14 03/15 03/16 03/17 03/18 03/19 Baseline Scenario Adverse Scenario Baseline Scenario Adverse Scenario Source: CNB Source: CNB 25

Macroeconomic developments, credit risk and stress scenario (III) The scenarios also included a marked rise in long-term interest rates, even in the deflation scenario. Alternative scenarios: 3M PRIBOR (%) 3 Alternative scenarios: 5Y CZK bond yield (v %) 6 5 2 1 4 3 2 1 0 0 03/13 03/14 03/15 03/16 03/17 03/18 03/19-1 03/13 03/14 03/15 03/16 03/17 03/18 03/19 Baseline Scenario Adverse Scenario Baseline Scenario Adverse Scenario Source: CNB Source: CNB 26

Bank stress test results confirm resilience of this sector In the Baseline Scenario three banks fall below the 8% level: these are small banks accounting for 1.8% of the sector s total assets, additional capital of CZK 0.5 billion is needed. In the Adverse Scenario seven banks would fall below the 8% threshold: they account for 7.5% of the sector s assets, additional capital of almost CZK 7 billion is needed. Additional sensitivity analyses: impairment of exposures to indebted EU countries, losses from operational risk. Capital ratios of the banking sector depending on scenarios (%) 20 18 16 14 12 10 8 6 4 03/13 03/14 03/15 03/16 03/17 03/18 03/19 Source: CNB, CNB calculation Baseline Scenario Adverse Scenario Adverse Scenario amplified by country risk Adverse Scenario amplified by operational risk 27

Detailed results of bank stress tests Impact of the alternative scenarios on the banking sector Baseline Scenario Adverse Scenario 2016 2017 2018 2019Q1 2016 2017 2018 2019Q1 Expected credit losses (minus sign for losses) CZK billions -17,1-20,7-23,1-6,7-45,8-77,2-78,5-17,5 % of assets -0,3-0,3-0,4-0,1-0,8-1,4-1,4-0,3 Profit/loss from market risks CZK billions 4,5-11,5-5,4-0,8-14,1-3,7-0,7 0,0 % of assets 0,1-0,2-0,1 0,0-0,2-0,1 0,0 0,0 Earnings for covering losses (adjusted operating profit) CZK billions 73,9 72,4 71,7 15,0 63,3 49,0 38,4 8,6 % of assets 1,2 1,2 1,1 0,2 1,1 0,9 0,7 0,2 Pre-tax profit/loss CZK billions 61,3 40,2 43,2 7,6 3,3-33,1-40,9-8,9 % of assets 1,0 0,6 0,7 0,1 0,1-0,6-0,7-0,2 Capital ratio at end of period in % Total 18,5 18,0 17,5 17,0 16,3 13,1 12,4 12,3 Tier 1 18,0 17,6 17,1 16,6 15,9 12,8 12,1 12,0 Capital injections CZK billions % of GDP No. of banks below 8% capital ratio Source: CNB, CNB calculation 0,5 6,6 0,01 0,1 3 7 28

Bank stress test results in event of elevated threshold The banking sector as a whole withstands the stress in the Adverse Scenario even with a more broadly defined minimum requirement. The threshold for passing the tests can be extended to include Pillar 2 requirements or a systemic risk buffer. The sum of the capital conservation buffer and the capital surplus creates a sufficient buffer for the absorption of the aggregate impacts of shocks. This applies to most banks. Impact of macro-stress tests (weighted average for sector as of end of 2015) 20 18 16 14 12 10 8 6 4 2 0 Overall Capital Requirement (OCR) 4,8 2,5 1,6 1,6 8,0 Combined Buffer Requirement (CBR) Total SREP Capital Requirement (TSCR) Impact of macro-stress tests Capital surplus Capital conservation buffer Systemic risk buffer Additional Pillar 2 requirements Pillar 1 requirements Relevant capital ratio Source: CNB Note: The illustration assumes a zero countercyclical buffer. 6,1 29

Results of bank balance-sheet liquidity tests The banking sector remains resilient to liquidity risk. The levels of short-term liquidity indicators are well above the 100% limits. The CNB s new macro-stress liquidity tests confirm the resilience of most banks to liquidity risk even at the oneyear horizon of the stress in the Adverse Scenario. The resilience is due to a strong client deposit base, and sufficient buffer comprised of high quality and liquid assets. Comparison of selected bank balance-sheet liquidity indicators (%) 1200 1000 800 600 400 200 0 Total LCR CNB liquidity test Large banks Mediumsized banks Small banks Building societies Source: CNB, CNB calculation Note.: The LCR is the ratio of the liquidity buffer to the net liquidity outflow of banks over a 30-day stress horizon as defined by EC Regulation 2015/61. The CNB liquidity test indicator is the ratio of the liquidity buffer to the net liquidity outflow of banks over a one-year stress horizon as defined by the CNB. 30

Stress test results for pension management companies The pension management companies sector remains sensitive to a fall in securities prices. The result is in line with the composition of the portfolio of transformed funds, where bonds sensitive to interest rate changes are dominant. In the event of a sharp rise in interest rates, pension management companies would have to increase capital. The considered shock generates a need for capital injections (totalling CZK 74.7 billion). However, the adverse effect fades in the case of long-term bond holdings. Pension Management Companies (PMC) stress test PMC Equity (as of 12/2015) Capital ratio (as of 12/2015) Change in TF asset value Interest rate risk Change in TF asset value Exchange rate risk Change in TF asset value Equity risk Change in TF asset value Property risk Total impact of risks on TF assets Profit of transformed funds PMC Equity (end of test) Capital ratio (end of test) Baseline Scenario Adverse Scenario mld. Kč 8,8 8,8 % 136,1 136,1 CZK bn -2,1-20,5 % assets -0,6-5,7 CZK bn 0,0 0,5 % assets 0,0 0,1 CZK bn -0,3-1,9 % assets -0,1-0,5 CZK bn 0,1-0,2 % assets 0,0 0,0 CZK bn -2,3-22,0 % assets -0,6-6,1 CZK bn 4,0 1,0 CZK bn 9,2 2,3 % 142,7 33,8 Capital injection CZK bn 0,0 4,7 Source: CNB, CNB calculation Note: TF = transformed funds of pension management companies 31

Insurance sector developments The CNB is not conducting its own stress test of insurance companies this year, as they are involved in EIOPA stress tests. The sector s profitability is stable and its capitalisation is sufficient. The solvency ratio exceeds three times the regulatory minimum. The non-life insurance market responded to the recovery in the Czech economy by growing. The life insurance market saw a decline; the move away from traditional products with guaranteed returns continues. Economic growth and the non-life insurance market in the Czech Republic (%) 8 6 4 2 0-2 -4-6 2004 2006 2008 2010 2012 2014 Year-on-year change in gross premiums written Annual nominal GDP growth Source: CNB, CZSO, Czech Insurance Association Note: The data on premiums written cover all insurance companies active in the Czech Republic, including branches of foreign insurance companies. 32

Sovereign exposures and public finance stress test Exposures to Czech government debt were again identified as systemically important this year (around CZK 680 billion at the end of 2015)... Banks with exposures exceeding 100% of eligible capital accounted for 68% of the total assets of the sector.... their riskiness (as measured by the ISR indicator) did not exceed the thresholds of 5% and 8% set in the internal methodology over the three-year horizon of the stress test. Czech public finance stress test Indicator of sovereign risk (ISR; v %) Adverse Scenario 2016 2017 2018 Critical limit 0,23 0,80 0,27 8,0 % (unconditional) 5,0 % (if confirmed by further analysis) In line with the CNB s internal methodology, no additional Pillar 2 capital requirement will be required for sovereign exposures. 33

Thematic articles in FSR 2015/2016 Could the Czech insurance sector be a source of systemic risk? The relationship between liquidity risk and credit risk in the CNB s liquidity stress tests The role of the leverage ratio in capital regulation of the banking sector. 34

Thank you for your attention Financial Stability Reports: http://www.cnb.cz/en/financial_stability/fs_reports/index.html Contact for the Financial Stability Department: E-mail: financial.stability@cnb.cz http://www.cnb.cz/en/financial_stability/