ANALYSIS

DANMARKS

NATIONALBANK 27 NOVEMBER 2019 — NO. 25

FINANCIAL STABILITY – 2ND HALF 2019 Lower excess capital adequacy for the banks

The largest credit institutions’ results remain high, underpinned by low loan impairment charges and CONTENTS income from mortgage loan refinancing. Several institutions have launched initiatives aimed at securing 2 SUMMARY future earnings. AND ASSESSMENT

4 PROSPECTS OF MORE The largest credit institutions have reduced excess MODERATE GROWTH AND PERSISTENTLY capital adequacy, and for a few it is low. The institutions LOW INTEREST RATES should reconsider their capital targets to ensure an 6 BANK EARNINGS appropriate distance between capitalisation and capital REMAIN HIGH

requirements. 13 A FEW INSTITUTIONS HAVE LOW EXCESS CAPITAL ADEQUACY The  largest banks generally have robust liquidity. Sen- 18 HOMEOWNERS OPT sitivity analyses of the banks’ liquidity are important for FOR FIXED RATE an understanding of the risks associated with liquidity MORTGAGE LOANS withdrawal in a crisis. 20 STRONG COMPETITION FOR CORPORATE CUSTOMERS

22 LIQUIDITY IS ROBUST, BUT SENSITIVE TO DEPOSIT WITHDRAWAL

27 APPENDIX

Negative Homeowners Intensified deposit rates opt for competitive for households will fixed rate mortgage loans pressure and declining have a limited impact and become more resilient margins for corporate on earnings to changes customers

Read more Read more Read more ANALYSIS — 2 FINANCIAL STABILITY – 2ND HALF 2019

Summary and assessment

Prospects of more moderate growth bank finds that the introduction of negative deposit and persistently low interest rates rates of -0.75 per cent for Danish households will Global economic growth is declining, and growth have a limited impact on the sector’s overall earn- expectations have been reduced, also for the euro ings. An interest rate of -0.75 per cent on household area. In September, the European , ECB, deposits exceeding kr. 750,000 would cause the lowered its deposit rate by 0.1 percentage point to sector’s earnings to rise by approximately kr. 1.1 -0.50 per cent, and the market expects even lower billion, disregarding depositors’ opportunities of re- interest rates in the coming years. At the same time, sponding to the negative interest rates. For instance, Danmarks Nationalbank lowered its certificate of households with both debt and bank deposits could deposit rate to -0.75 per cent. The accommodative respond by reducing deposits by means of debt has led to a decrease in short-term servicing. money market interest rates and substantial declines in the yields on Danish mortgage and government Institutions should ensure an appropriate distance bonds, which reached the lowest levels ever in the between capital targets and capital requirements autumn of 2019. The systemic credit institutions’ capitalisation has been stable in recent years, but increasing capital Results remain high, but core earnings are falling requirements have reduced their excess capital The systemic credit institutions’ earnings remain adequacy. Danmarks Nationalbank finds that a few high, underpinned by low loan impairment charges. of the systemic institutions should reconsider their However, the long period of negative interest rates capital targets to ensure an appropriate distance has entailed lower core earnings for the institutions. between capitalisation and capital requirements. The Net fee income and income from administration need to reconsider capital targets is underscored by margins have receded a little in recent years, mean- Danmarks Nationalbank’s most recent stress test of ing that they are making up for the decline in net the institutions’ solvency. It shows that a few of the interest income to a lower degree than previously. systemic banks will fall short of their capital buf­ The large wave of mortgage refinancing in 2019, es- fer requirements in a severe stress scenario even if pecially in the 2nd half, will have a positive impact on the countercyclical capital buffer is assumed to be net fee income, but as a one-off effect only. Weaker released. earnings will reduce the institutions’ first line of de- fence against losses. It is important that the institu- is one of the EU member states tions continue to enhance efficiency and reduce their where the new Basel rules are expected costs to offset falling core earnings. to have the strongest impact The forthcoming implementation of the 2017 Basel Households have so far been exempt reform is also a reason for the systemic institutions from negative deposit rates to reconsider their capital planning. Like institutions For households, the banks have not reduced inte­r­ in other Northern European countries, Danish insti- est rates to the same extent as they have done for tutions are among those that should be expected to corporate customers. As a result, the banks are, on see the strongest increases in capital requirements. average, paying zero or positive deposit rates to The increases can predominantly be attributed households, while placing excess liquidity in the to the introduction of the output floor, which will market at negative interest rates. Together with the impose a lower limit for risk-weighted exposures rising customer funding surplus, this puts increas- calculated on the basis of historical losses. ing downward pressure on the banks’ net interest income. MREL compliance reached as at 1 July 2019 The systemic institutions all met the minimum Negative deposit rates for households requirement for eligible liabilities, MREL, on 1 July will have a limited impact on earnings 2019, when it became effective. However, the excess Several banks have announced that they will intro­ capital adequacy relative to the new requirement is duce negative deposit rates for households on limited, and findings from Danmarks Nationalbank’s deposits exceeding kr. 750,000. Danmarks National- stress test show that the systemic institutions will ANALYSIS — DANMARKS NATIONALBANK 3 FINANCIAL STABILITY – 2ND HALF 2019

have a substantial need to issue new MREL-eligible The banks’ liquidity is robust, instruments in a stress scenario. Danmarks Nation- but sensitive to deposit withdrawal albank expects the institutions to continue building Danmarks Nationalbank has developed a tool for up MREL-eligible instruments in the coming years. assessing bank liquidity sensitivity. The sensitivity Given that periods of no or limited capital market analysis shows that the liquidity of the systemic access may arise, the institutions should focus on banks is generally robust to a market downturn the length and maturity profiles of their issuances where the banks cannot expect to be able to renew to avoid strong concentration of maturing issuances existing loans and issuances when they mature. The and breach of the MREL. sensitivity analysis also shows that the systemic banks’ liquidity position is more sensitive to an insti- Homeowners opt for fixed rate mortgage loans, tution-specific shock, for instance with doubt about becoming increasingly resilient to changes whether an individual bank will survive and with in interest rates depositors withdrawing funds. It is important that Overall growth in lending to households is driven the banks themselves have a deep understanding by fixed rate mortgage loans. Fixed rate mortgage of the sensitivity to various risks affecting liquidity loans now account for 46 per cent of mortgage withdrawal in a crisis. lending, compared with 33 per cent at the beginning of 2014. Consequently, households have become more resilient to changes in interest rates on their debt. Fixed rate mortgage lending is increasing in all municipalities, but most in and environs and around the large cities. Both banks and mort- gage credit institutions report lower risk appetite for lending to households in recent quarters. Risk appetite has declined due to, inter alia, changes to good practice for lending for housing purposes and the guidelines on prudent credit assessment when granting housing loans in growth areas.

Intensified competitive pressure and falling interest margins for corporate lending Growth in lending to corporate customers continues at a moderate pace. Credit standards for corporate customers were more or less unchanged in the 3rd quarter of 2019, but both banks and mortgage ­credit institutions still report that competitor behaviour has contributed to their easing of credit standards. The competitive pressure is primarily reflected in the banks’ interest margin, which continues to decline. It is important that the institutions maintain a sound credit quality and that the interest margin reflects the risk on lending. Since losses do not materialise until the economy reverses, the institutions should assess firms’ resilience over the entire business cycle. ANALYSIS — DANMARKS NATIONALBANK 4 FINANCIAL STABILITY – 2ND HALF 2019

Prospects of more moderate growth and persistently low interest rates

A more moderate growth outlook Global growth is declining. The latest forecasts from Downward adjustment of expected Chart 1 the International Monetary Fund, IMF, predict global growth in the global economy, economic growth, but at a moderate level. The IMF the euro area and Denmark has reduced its global growth estimate, now expect- Real GDP growth, ing global economic growth of 2.5 per cent for 2019, per cent year-on-year relative to 2.7 per cent in April. This is the lowest 3.0 level since 2012. The downward revision of the IMF 2.5 growth estimates is motivated by, inter alia, the trade 2.0 conflict between the USA and China and a general 1.5 slowdown. However, the forecast for the US econ­ 1.0 omy has been adjusted slightly upwards since April, 0.5 but the expectation for 2020 is still for lower growth 0.0 compared with 2019. 2019 2020 2019 2020 2019 2020 2019 2020 World USA Euro area Denmark

Expected growth in the euro area has also been ad- Autumn 2019 forecast Spring 2019 forecast justed downwards since April, primarily due to weak industrial production, which has fallen in Germany, among other countries. Similarly, Danmarks Nation- Note: The chart shows the IMF forecast for the world based on albank’s most recent projection indicates declining exchange rates and the IMF forecasts for the USA and the euro area. The forecast for Denmark is from Danmarks growth in the Danish economy in 2020, cf. Chart 1. Nationalbank’s projections from March 2019 and Sep- tember 2019. Source: IMF and Danmarks Nationalbank. Lower monetary policy interest rates in response to slowdown In September, the , ECB, lowered its deposit rate by 0.1 percentage point to The market expects Chart 2 -0.50 per cent. At the same time, the ECB introduced even lower interest rates a two-tier interest rate system allowing banks to deposit a part of their liquidity at 0 per cent. The Per cent ECB also announced that it would resume its asset 3.0 purchase programme. The economic stimulus in the 2.5 Expectation in FOMC form of a lower key interest rate is implemented 2.0 after the ECB’s downward adjustment of inflation 1.5 expectations against the background of mounting USA 1.0 concerns about the global macroeconomic situation. 25 November 2019 0.5 The market expects even lower interest rates in the 0.0 coming years, cf. Chart 2. Euro area -0.5 25 November 2019 The , FED, has also lowered the fed -1.0 funds target rate. This has led to a fall in market 2018 2019 2020 2021 expectations of future interest rates. Note: Fed funds target rate for the USA, the ECB’s deposit Interest rate spike in the secured US money market rate for the euro area. The dotted lines denote market expectations based on futures prices. The most recent On 17 September 2019, the price of a secured over- observations are from 25 November 2019. night dollar loan, i.e. the repo rate, rose to around Source: Federal Reserve, ScanRate RIO and Macrobond. 9 per cent during one day, cf. Chart 3. On that day and the subsequent days, the FED operated in the market with measures to reduce the repo rate, thus aiming to keep the rate for unsecured ANALYSIS — DANMARKS NATIONALBANK 5 FINANCIAL STABILITY – 2ND HALF 2019

borrowing within 2-2.25 per cent, which was the target. The repo rate hike was presumably a conse- Strong increase in interest rates Chart 3 quence of several concurrent events resulting in a in the secured US money market dollar shortage. Rate, per cent 9 In the wake of the strong repo rate fluctuations, the 8 FED announced that it would resume purchasing 7 government bonds, mainly in order to increase the 6 Secured overnight financing rate volume of dollars. 5 4 3 Danmarks Nationalbank has reduced 2 the key monetary policy interest rate 1 The fixed exchange rate policy means that Denmark’s 0 monetary policy is aimed at keeping the stable 29 Jul 07 Oct 17 Oct 27 Oct 07 Sep 17 Sep 27 Sep 08 Aug 18 Aug 28 Aug against the euro. That is why Danmarks Nation- 06 Nov albank mirrored the ECB and lowered the rate of interest on certificates of deposit by 10 basis points Note: The chart shows the 99th percentile of the secured over- to -0.75 per cent from 13 September 2019, close to night financing rate. The most recent observation is from 6 November 2019. an all-time low. The lending, current account and Source: of New York. discount rates are unchanged. At the moment, the monetary policy counterparties need to deposit substantial funds with Danmarks Nationalbank, and Chart 4 in this situation the monetary policy deposit rates Banks’ portfolios of negative yield bonds have increased determine the money market rates and the exchange rate of the krone. Net position, Kr. billion 500 Falling interest rates affect 400 the return on banks’ bond portfolios Yield to maturity > 0 The accommodative monetary policy has led to a 300 0 > decrease in short-term money market interest rates Yield to maturity > -0,25 and substantial declines in the yields on Danish 200 mortgage and government bonds. In the autumn of -0,25 > Yield to maturity > -0,5 2019, yields on mortgage and government bonds 100 were the lowest ever. The very low level of interest -0,5 > Yield to maturity rates affects yields on the banks’ bond portfolios 0 held for liquidity management purposes. The volume Sep 16 Mar 17 Sep 17 Mar 18 Sep 18 Mar 19 Sep 19 of bonds with a yield to maturity of less than -0.5 per cent has thus risen considerably during 2019, cf. Note: The chart shows the return on systemic banks’ bond Chart 4. portfolios by yield to maturity, i.e. the return on a bond if held to maturity. The most recent observation is from September 2019. Bank equities have lost value, Source: Danmarks Nationalbank and Bloomberg. and the price/book ratio is below 1 The price/book ratio for a bank’s equity may provide an indication of investors’ confidence in the bank’s ability to generate value for the shareholders. A level below 1 has usually indicated lower investor confi- dence in the bank’s ability to perform well in future, or that the asset value reflects the book value.

The average price/book ratio for Danish systemic banks has been below 1 since the end of 2018. For the Danish systemic banks, this ratio has thus moved from being lower than the level for comparable ANALYSIS — DANMARKS NATIONALBANK 6 FINANCIAL STABILITY – 2ND HALF 2019

Nordic banks in the period 2015-18 to being consid- erably lower, cf. Chart 5. The price/book ratio is lower Chart 5 for Danish banks than The falling price/book ratio is attributable to the for comparable Nordic banks development in return on equity, among other Price/book ratio factors. By comparison to European banks, Danish 1.8 banks posted a high return on equity in 2016 and Nordic banks 2017, but the return declined in 2018 and early 2019 1.6 to a substantially lower level than that of the Nordic 1.4 banks. Since 2012, the Nordic banks have managed Danish banks 1.2 to maintain a high return on equity. 1.0

New reference rates 0.8 In October, the reference rate for unsecured over- European banks 0.6 night interbank loans in the euro area, EONIA, was 0.4 replaced by the new euro short-term rate, €STR, as 2012 2013 2014 2015 2016 2017 2018 2019 reference rate. €STR is based on interbank trans­ actions and transactions between banks and other financial institutions. The ECB publishes the refer- Note: Danish banks comprise , Jyske Bank, Sydbank and Spar Nord. Nordic banks consist of DNB, ence rate daily. The transition to a new interest rate Svenska Handelsbanken, Swedbank, Nordea and SEB. can be a huge task for financial corporations, due European banks are based on the STOXX Europe 600 Banks index. Unweighted averages of weekly observa- to EONIA’s very extensive use as a reference rate tions. The most recent observation is from 22 November on a number of financial products and in risk and 2019. Source: Bloomberg. valu­ation models. This transition is an element of a ­major international effort to replace traditional inter­bank interest rates, so-called IBORs, with alter­ native transaction-based reference rates. Similar Earnings are buoyed up Chart 6 work on an overnight reference rate is being per- by low loan impairment charges formed in Denmark under the auspices of Finance Denmark. The introduction of a Danish transac- Kr. billion 30 tion-based reference rate will increase the confi- Profit before tax dence in that rate and will be in accordance with in- 25 ternational developments in the area. With the aim 20 of clarification and of strengthening confidence in 15 reference rates in Denmark, the new reference rate 10 Loan impairment should be implemented rapidly and the existing charges 5 one should be phased out. It is also important to 0 find solutions concerning reference rates for longer maturities and for migrating existing contracts from -5 old to new reference rates. -10 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19

Note: Six-month data for systemic credit institutions. The result is before tax and adjusted for goodwill impairment Bank earnings remain high charges. The most recent observation is from the 1st half of 2019. Source: Danmarks Nationalbank.

Results remain high, but core earnings are falling The systemic credit institutions’ earnings remain high, also in the 3rd quarter of 2019. The return on equity for these institutions was 8.4 per cent p.a. in the 1st half of 2019, corresponding to the level at end-2015, and the interim result is still well above the results in the post-crisis period, cf. Chart 6. ANALYSIS — DANMARKS NATIONALBANK 7 FINANCIAL STABILITY – 2ND HALF 2019

Earnings have been buoyed up by low loan impair- ment charges for a prolonged period, cf. Chart 6. The Core earnings have Chart 7 low loan impairment charges should be viewed in declined in recent years the light of the favourable cyclical situation and the low level of interest rates, which makes it easier for Kr. billion Kr. billion households and firms to service their debts. Earnings 25 3,0 Systemic credit institutions have also been supported by recent years’ positive value adjustments. 20 2,5 Tusinde

However, the long period of negative interest rates 15 2,0 has entailed falling core earnings, cf. Chart 7. Core earnings consist mainly of net interest and net fee in- 10 1,5 come less costs and are a measure of the institutions’ 5 1,0 ability to make a profit on their core business. The non-systemic banks’ core earnings increased in the Non-systemic banks (right-hand axis) 0 0,5 1st half of 2019 due to such factors as rising net fee 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 income. Weaker earnings will reduce the institutions’ first line of defence against losses. Note: Six-month data for systemic credit institutions and non-systemic banks. Core earnings are defined as profit Net interest income accounts for before tax less dividend from equity investments, value a smaller share of earnings adjustments, loan impairment charges and income from investments. The systemic credit institutions’ core The systemic credit institutions’ declining core earn- earnings have been adjusted for goodwill impairment ings are mostly attributable to falling net interest charges. The most recent observation is from the 1st half of 2019. income, cf. Chart 8. At the same time, net fee income Source: Danmarks Nationalbank. has receded a little in recent years, while income from administration margins is almost unchanged. Net fee income and income from administration margins are thus making up for the decline in net interest income to a lower degree than previously.

Sustained fall in net interest income Chart 8 Kr. billion Systemic credit institutions Non-systemic banks 30

Kr. billion Kr. billion 3025 3.5

Net interest income 2520 3.0

20 2.5 15

15 Administration2.0 margin 10 10 1.5

5 5 1.0 Net fee income

0 0.5 05 0506 07 0806 09 100711 120813 14 0915 16 1710 18 1911 12 1305 06 0714 08 091510 11 1612 13 141715 161817 18 1919

Net interest income, excl. administration margin Net fee income Administration margin

Note: Six-month data for systemic credit institutions and non-systemic banks. The most recent observation is from the 1st half of 2019. Source: Danmarks Nationalbank. ANALYSIS — DANMARKS NATIONALBANK 8 FINANCIAL STABILITY – 2ND HALF 2019

For the non-systemic banks, net interest income has also declined in recent years, but core earnings have Average administration Chart 9 been buoyed up by higher net fee income, cf. Chart margins are receding 8. Also in the 3rd quarter of 2019, core earnings were Per cent buoyed up by higher net fee income. 1.0 Households 0.9 The development in administration margins and net fees is driven by e.g. mortgage changes 0.8

The systemic credit institutions’ income from admin- 0.7 istration margins has been almost unchanged since Corporates 2017 despite continually increasing mortgage lend- 0.6 ing. The flattening of income from administration 0.5 margins reflects a drop in average administration 0.4 margins as more borrowers convert their loans to safer loan types with lower administration margins, 0.3 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 cf. Chart 9. In recent years, the institutions changed their administration margin structure, so that loans with deferred amortisation and short fixed-interest Note: Average administration margins for outstanding domes- periods became relatively more expensive. tic mortgage lending in Danish kroner. The most recent observation is from September 2019. Source: Danmarks Nationalbank. For the non-systemic banks participating in the Total­ kredit partnership, income from mortgage lending appears as fee income. Recent years’ strong increas- es in non-systemic banks’ net fee income are driven, est-bearing assets. As market rates have fallen, inter- inter alia, by higher fee income from intermediation est income from lending has made up an increasing of mortgage loans. share of the institutions’ interest income, cf. Chart 10.

In contrast, net fee income for the systemic credit in- Historically, the return on interest-bearing assets stitutions taken as one has been declining since end- has been higher for the non-systemic banks than for 2017. A contributing factor is the systemic banks’ the systemic banks. The higher return reflects higher lower fee income from securities trading. In addition, average lending rates for households and corporate fee income will fall as more and more homeowners customers, cf. Chart 11, and should be viewed in con- opt for mortgage loans with longer fixed-interest junction with, inter alia, the credit quality of the loan periods, since a brokerage fee is payable for each portfolio.2 refinancing. The spread between the lending rate and the money The large volume of refinancing of mortgage loans in market interest rate has narrowed in recent years for 20191 has a positive impact on net fee income in 2019, both systemic and non-systemic banks, driven by the especially in the 2nd half. But this is a one-off effect competitive situation, among other factors.3 which will only temporarily boost core earnings. Banks have adapted deposit rates for Interest income has declined due corporate customers to the negative market rates to lower lending rates Falling market rates have also entailed lower interest The banks’ interest income has been falling for a costs, cf. Chart 12. The systemic banks’ interest costs prolonged period despite the slight increase in inter- are affected more by market rates, since they also

1 See the section “Homeowners opt for fixed rate mortgage loans”. 2 See the section “Homeowners opt for fixed rate mortgage loans” and Danmarks Nationalbank, Prospects of lower earnings and higher capital requirements for banks, Danmarks Nationalbank Analysis (Financial stability, 1st Half 2019), No. 11, May 2019.

3 See the section “Homeowners opt for fixed rate mortgage loans”. ANALYSIS — DANMARKS NATIONALBANK 9 FINANCIAL STABILITY – 2ND HALF 2019

Interest income has declined despite the slight increase in interest-bearing assets Chart 10 Mia. kr. Mia. kr. Systemic banks Non-systemic banks 120 3.000 Rentebærende aktiver (højre akse) Kr. billion Kr. billion Kr. billion Kr. billion 100 2.500 120 3,000 10 250

10080 2,500 2.000 8 200

80 2,000 60 Obligationer 1.500 6 150 60 1,500 40 Tilgodehavender 1.000 4 100 40 1,000 20 500 2 50 20 500

0 0 0 0 0 0 050506 07 0806 09 100711 12 1308 14 15 1609 17 Udlån181019 11 12 1305 06 1407 08 091510 111612 13 141715 16 1817 18 1919

Lending Claims on credit institutions and central banks Bonds Interest-bearing assets (right-hand axis)

Note: Six-month data for systemic and non-systemic banks. Interest-bearing assets are defined as the sum of the following assets: claims on credit institutions and central banks, lending and other receiveables at fair value and amortised cost, and bonds at fair value and amort­ ised cost. The most recent observation is from the 1st half of 2019. Source: Danmarks Nationalbank and own calculations.

Lending rates have decreased faster Chart 11 issue bonds as a source of funding, in addition to than deposit rates in recent years deposits. Interest costs have fallen, despite recent years’ slight increase in interest-bearing liabilities. Per cent 10 Practically all of this increase is driven by deposits. Non-systemic banks 8 The systemic banks have lower average deposit rates for corporate customers than the non-systemic 6 banks, and the former have adapted deposit rates to 4 money market interest rate developments to a high- er degree, cf. Chart 13. In April 2019, 63 per cent of 2 Systemic banks the systemic banks’ corporate deposits were subject 0 to negative interest rates, compared with 23 per cent T/N money market interest rate for the non-systemic banks, cf. Chart 14. Some of this -2 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 variation is due to the systemic and non-systemic banks having different business models. The extent of negative deposit rates for corporate customers Note: Lending rates comprise loans in all currencies vis-à-vis varies substantially across the individual banks, but Danish counterparties (excluding bad loans). A few non-systemic banks are not included for the whole there is a general trend towards increased use of ­period, as they have not reported to Danmarks National- negative interest rates. bank’s interest rate statistics. The most recent obser- vation is from September 2019 for lending rates and October 2019 for the T/N money market interest rate. Households have so far been exempt Source: Danmarks Nationalbank and Nasdaq OMX Nordic. from negative deposit rates For households, the banks have not reduced de- posit rates to the same extent as they have done for corporate customers, cf. Chart 15. As a result, the banks are, on average, paying positive deposit rates to households, while placing excess liquidity in the ANALYSIS — DANMARKS NATIONALBANK 10 FINANCIAL STABILITY – 2ND HALF 2019

Interest costs have declined despite the slight increase in interest-bearing liabilities Chart 12 Mia. kr. Mia. kr. Systemic banks Non-systemic banks 60 3.000 Kr. billion Kr. billion RentebærendeKr. billion Kr. billion passiver (højre akse) 5060 3,000 4 4002.500 Tusinde

50 2,500 40 2.000 3 300 Tusinde Tusinde 40 2,000 30 Indlån 1.500 30 1,500 2 200 20 1.000 20 Kreditinstitutter1,000

10 1 100500 10 500 Efterstillet kapitalindskud

0 0 0 05 06 07 08 09 10 0 11 12 0 13 14 15 16 17 18 19 0 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 Udstedte obligationer Bonds issued Credit institutions and central banks Deposits Subordinated debt Interest-bearing liabilities (right-hand axis)

Note: Six-month data for systemic and non-systemic banks. Interest-bearing liabilities are defined as the sum of the following liabilities: debt to credit institutions and central banks, deposits and other debt (including deposits in pooling schemes), bonds issued at fair value and amortised cost and subordinated capital. The most recent observation is from the 1st half of 2019. Source: Danmarks Nationalbank and own calculations.

Negative deposit rates Chart 13 The systemic banks have adapted Chart 14 for corporate customers deposit rates to a higher degree

Per cent Per cent 6 100 80 5 60 4 40

3 20

0 2 Apr 17 Apr 19 Apr 17 Apr 19 Apr 17 Apr 19 Apr 17 Apr 19 Non-systemic banks 1 Systemic Non- Systemic Non- Systemic banks systemic systemic 0 Households Corporates T/N money market interest rate -1 Negative rate Zero rate Positive rate 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19

Note: Shares of deposits (excluding deposits in pooling Note: Deposit rates comprise deposits in all currencies from schemes), as percentages of the average balance sheet, Danish corporate customers. A few non-systemic banks invested at positive, zero and negative interest rates, re- are not included for the whole period, as they have spectively, for Danish customers. The non-systemic banks not reported to Danmarks Nationalbank’s interest rate exclude Alm. Brand Bank. statistics. The most recent observation is from Septem- Source: Danmarks Nationalbank and own calculations. ber 2019 for deposit rates and October 2019 for the T/N money market interest rate. Source: Danmarks Nationalbank, Nasdaq OMX Nordic and DFBF. ANALYSIS — DANMARKS NATIONALBANK 11 FINANCIAL STABILITY – 2ND HALF 2019

­market at negative interest rates. In step with the rising customer funding surplus, this puts increas- Household deposit rates Chart 15 ing downward pressure on the banks’ net interest are close to zero income. All else equal, this combination will hit the Per cent non-systemic banks the hardest, as their lending ac- 6 tivities are financed by deposits to a higher degree. 5

The non-systemic banks have adapted household 4 deposit rates more slowly, but the deposit rate gap 3 between systemic and non-systemic banks has nar- 2 rowed in recent years. This development should be Non-systemic banks viewed in the light of the non-systemic banks’ more 1 extensive use of zero interest rates for households, Systemic banks as the share of household deposits at zero interest 0 T/N money market interest rate rate has risen from 35 per cent in 2017 to almost -1 50 per cent in 2019, cf. Chart 14. The non-systemic 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 banks’ slower adaptation of household deposit rates may be related to their relatively more pronounced Note: Deposit rates comprise deposits in all currencies from dependence on deposit financing, so they are more Danish households. A few non-systemic banks are not included for the whole period, as they have not reported careful about retaining customers. to Danmarks Nationalbank’s interest rate statistics. The most recent observation is from September 2019 for de- posit rates and October 2019 for the T/N money market It should be noted that although the deposit rate interest rate. has dropped to zero or close to zero for households, Source: Danmarks Nationalbank, Nasdaq OMX Nordic and DFBF. long-term interest rates have declined even more – albeit from a higher level. Accordingly, the spread between these two interest rates has narrowed. This Low deposit rates for Danish Chart 16 implies a smaller loss of interest on bank deposits household customers rather than holding bonds or servicing debt. All else equal, this makes bank deposits cheaper and more Average 0.07 per cent Ordinary positive attractive today compared with previously.4 rate deposits

0.23 Ordinary per cent The systemic and non-systemic banks’ average zero rate deposits deposit rates are to some extent buoyed up by the Fixed rate Average 0.25 deposits interest rates on so-called lending-related deposits, 0 per cent per cent 0.30 cf. Chart 16. When a bank grants a mortgage-like per cent bank loan, the customer is often also offered a de- 2.22 per cent posit account which accrues interest at the lending Return Lending- dependent related rate. Since this lending rate is normally substantially rate of deposits higher than the average deposit rate, this type of interest deposit will push up the calculated average deposit Pooling schemes rate. Even when adjusted for lending-related de­ posits, deposit rates vary between systemic and Note: Average deposit rates in Danish kroner for “Employees, non-systemic banks. pensioners, etc.” in Denmark. Data is for April 2019. Source: Danmarks Nationalbank.

Negative deposit rates for households will have a limited impact on earnings Several banks have announced that they will intro- duce negative deposit rates for households in the

4 See Danmarks Nationalbank, Low interest rates boost bank deposits, Danmarks Nationalbank Analysis, No. 9, July 2018 (link). ANALYSIS — DANMARKS NATIONALBANK 12 FINANCIAL STABILITY – 2ND HALF 2019

coming months. Danmarks Nationalbank finds that the introduction of negative deposit rates of -0.75 Not all household deposits can be Chart 17 per cent for Danish households will have a limited subject to negative interest rates impact on the sector’s earnings overall at the banks’ Kr. billion current thresholds for which deposits are affected by 1,200 negative deposit rates. 1,000 Other deposits Should the banks taken as one introduce negative (Kr. 236 billion) 800 deposit rates for households on all deposits irre- spective of size, deposits of approximately kr. 790 600 billion would be affected by negative interest rates, Deposits 400 with potentially cf. Chart 17. The remaining deposits of around kr. negative rates 240 billion include e.g. deposits invested by the bank (Kr. 786 billion) 200 in pooling schemes on the customer’s behalf, where the interest rate depends on the return on the pool- 0 ing scheme.

Note: Deposits with potentially negative rates consist of Most people have modest bank deposits. For ap- demand deposits excluding lending-related deposits (kr. proximately 70 per cent of the just under 5 million 692 billion), cash deposits in pension savings accounts people with deposit accounts, deposits amount to (kr. 55 billion), deposits redeemable at notice (kr. 25 billion) and time deposit excluding special deposit types less than kr. 100,000. By comparison, around 178,000 (kr. 14 billion). Cash deposits stated at end-2018. Lend- persons have bank deposits exceeding kr. 750,000, ing-related deposits stated at April 2019. Deposits stated at September 2019. 5 totalling around kr. 280 billion, cf. Chart 18. Source: Danmarks Nationalbank, Danish Financial Supervisory Authority and own calculations. An interest rate of -0.75 per cent on household deposits exceeding kr. 750,000 would cause the sec- tor’s earnings to rise by approximately kr. 1.1 billion, Household deposits by size Chart 18 disregarding depositors’ opportunities of respond- ing to the negative interest rates. This corresponds Kr. billion to 3.7 per cent of the banks’ total earnings of just 300 over kr. 30 billion in 2018. 250

200 The total effect on earnings of introducing negative deposit rates for households depends on the thresh- 150 old at which the negative deposit rates kick in. The 100 more deposits that are affected by negative deposit 50 rates, the higher the impact will be on the sector’s 0 earnings as a whole. A negative deposit rate on Kr. Kr. Kr. Kr. > Kr. deposits exceeding e.g. kr. 200,000 would result in a 0.1- 200,000- 500,000- 750,000- 7,500,000 200,000 500,000 750,000 7,500,000 positive annual effect on earnings of approximately kr. 2.8 billion, cf. Chart 19. This corresponds to 9.2 per cent of the banks’ total earnings in 2018. Note: The chart shows total deposits broken down by deposit size. Only total deposits per person can be calculated. Deposits can be distributed on several banks. This calculation covers 4.8 million persons. Register data from Households are likely to respond end-2018. if affected by negative interest rates Source: Statistics Denmark and own calculations. The presented forecasts of the effect of negative deposit rates on the sector’s earnings should be

5 Given that negative interest rates do not apply to the first kr. 750,000 of the deposit, deposits totalling around kr. 147 billion will accrue negative interest rates if all banks set the limit at kr. 750,000. ANALYSIS — DANMARKS NATIONALBANK 13 FINANCIAL STABILITY – 2ND HALF 2019

regarded as high-side forecasts. One reason is that the effect is overestimated for customers who have The effect on earnings of introducing Chart 19 spread their deposits across several banks. The negative deposit rates for households effect calculations do not take derived behavioural depends on the threshold effects in individual households into account either. Kr. billion 3 An example of a possible behavioural effect is that households with both debt and bank deposits could reduce deposits by means of debt servicing. More- 2 over, negative deposit rates on demand deposits or cash pension savings can be avoided by invest- 2.8 ing the funds in e.g. pooling schemes or securities. Finally, households can withdraw their deposits and 1 1.6 hold cash. Negative deposit rates may generate 1.1 higher fee income for the banks if the negative de- 0.1 posit rates induce customers to move a part of their 0 Kr. 7,500,000 Kr. 750,000 Kr. 500,000 Kr. 200,000 deposits to investment products. The behavioural effects of introducing negative deposit rates for households are described in more detail in Box 1. Note: The estimated effects on earnings at different limits of amounts are based on register data. The calculation does not take into account that deposits may be distrib- Rising costs reduce core earnings uted on several banks. The most recent observation is from end-2018. The banks’ costs have risen in recent years, primarily Source: Statistics Denmark and own calculations. due to higher staff costs and administrative expenses, cf. Chart 20. Despite the increased focus on costs, no substantial savings have been obtained yet, which can be attributed to, inter alia, rising costs for implemen- High costs reduce core earnings Chart 20 tation of regulatory and compliance measures. Kr. billion Kr. billion Given the rising costs and declining core earnings, 25 10 it is important that the institutions continue to focus on enhancing efficiency and reducing costs to offset 20 8 Tusinde falling core earnings. This applies even though in- Costs systemic credit institutions vestments in automation and digitalisation will only 15 6 contribute to efficiency gains in the longer term. 10 4

5 2 Costs non-systemic banks A few institutions have low excess (right-hand axis) 0 0 capital adequacy 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19

Note: Six-month data for systemic credit institutions and Reduced excess capital adequacy non-systemic banks. Costs are defined as the sum of staff The systemic credit institutions’ capitalisation has costs and administrative expenses, depreciation and amortisation of intangible and tangible assets and other been stable in recent years, cf. Chart 21. At the same operating costs. The systemic credit institutions’ costs time, the capital requirements that may vary over have been adjusted for goodwill impairment charges. The most recent observation is from the 1st half of 2019. time, i.e. the Pillar II requirement and the counter­ Source: Danmarks Nationalbank and own calculations. cyclical capital buffer, have increased. This implies a reduction of excess capital adequacy, i.e. less dis- tance between the institutions’ actual capital ratios and total capital requirements.

There is substantial variation in excess capital ad- equacy across the systemic credit institutions, and ANALYSIS — DANMARKS NATIONALBANK 14 FINANCIAL STABILITY – 2ND HALF 2019

How do households respond Box 1 to negative deposit rates?

Household deposit rates have fallen over a prolonged Household-level register data shows that many households ­period and are now zero or just above zero, cf. Charts 15 have both considerable deposits and substantial debt.1 This and 16. The falling deposit rates do not appear to have re- is not a new phenomenon. A rational approach for liquid- duced household deposits. Conversely, deposits have risen ity-preferring households could be to have both debt and faster than justified by the transaction requirement, e.g. bank deposits. Households can be hesitant to reduce debt measured by domestic demand. The increase in deposits if they are uncertain about the possibility of obtaining credit can be attributed to the lower spread between bond yields later when they need it. and deposit rates, which has made bank deposits more attractive. Calculations based on register data show that debt reduc- tion can reduce total household deposits by approximate- Negative deposit rates may be the turning point for when ly kr. 160 billion. This result is calculated assuming that households begin to reduce their deposits. Having to pay households with mortgage debt reduce their deposits to for holding money in the bank, i.e. a negative return, could 2 months’ income after tax.2 People with higher incomes have a psychological effect. The households’ response will thus have higher bank deposits than those with lower depends very much on their return and liquidity preferences. incomes. For an average household, 2 months’ income after Negative interest rates could change this, and the behaviour tax corresponds to around kr. 58,000. Under the above as- may show non-linearity when interest rates fall below zero. sumptions, 20 per cent of the households will reduce debt. The results are not particularly sensitive to assumptions of Households can reduce their deposits or avoid negative 1, 2 or 3 months’ income. If the deposit is instead reduced to interest rates in several ways: (1) They can reduce debt. (2) 1 month’s income, total deposits can potentially be reduced They can opt for fixed deposits which still carry positive by around kr. 190 billion, while the figure for 3 months’ interest rates, cf. Chart 16, but that would be at the expense income is around kr. 140 billion. of liquidity. (3) They can invest in e.g. pooling schemes or securities, which are also less liquid assets. (4) They can At the end of 2018, household deposits totalled kr. 990 avoid negative interest rates by spreading deposits across billion, while total mortgage debt and bank debt totalled several banks, if the institutions set a threshold for when around kr. 2,400 billion. A reduction of total deposits (and deposits are affected by negative interest rates. (5) They debt) by around kr. 160 billion thus corresponds to approxi­ can withdraw their deposits and hold more cash. However, mately 16 per cent lower deposits and around 7 per cent holding large amounts of cash is rather risky, not to mention lower debt.3 The share of deposits that can be reduced by the costs of safe storage. The first effect, debt reduction, is servicing debt has been relatively constant over the last 20 elaborated below. years. The constant share indicates that households prefer liquid funds. If a positive return is preferred over liquidity, Household deposits can potentially household behaviour may potentially shift when deposit be reduced by reducing debt rates become negative. Household deposits have risen slightly faster than house- hold debt in recent years. Household deposits and debt are driven by the same underlying factors – domestic demand, property prices and interest rates, among others – and tend to mirror each other over time.

1. In the register data-based calculation, bank deposits are primarily demand deposits. Tax-privileged deposit types (e.g. children’s savings and pension savings) are not included.

2. The calculation does not assume that households reduce their bank debt.

3. A previous Danmarks Nationalbank analysis indicates that money demand will be reduced by around 5 percentage points if deposit rates fall by 1 percentage point, cf. Erik Haller Pedersen and Jonas Ladegaard Hensch, Low interest rates boost bank deposits, Danmarks Nationalbank Analysis, No. 9, July 2018 (link). A decline in the deposit rate relative to the bond yield will increase the loss of interest on holding bank de­ posits rather than bonds or servicing debt, which will reduce deposits according to this analysis. ANALYSIS — DANMARKS NATIONALBANK 15 FINANCIAL STABILITY – 2ND HALF 2019

The distance between own funds Chart 21 Considerable variation in excess Chart 22 and has been capital adequacy across institutions reduced in systemic institutions Excess capital adequacy, per cent of risk-weighted exposures Per cent of riskweighted exposures 10 25 8 20 Excess capital Pct. af REA Total own funds adequacy 6 25 15 20 4 15 10 2 10 5 0 5 0 Bank Jyske

3. 4. 1. 2. 3. 4. 1. 2. 3. 4. 1. 2. 3. Bank Kredit Danske Nordea Bank

kvt. kvt. kvt. kvt. kvt. kvt. kvt. kvt. kvt. kvt. kvt. kvt. kvt. Sydbank

0 Nykredit Realkredit Spar Nord Q316 Q416 Q117 Q217 Q317 17Q4 18Q1 18Q2 18Q3 18Q4 19Q1 19Q2 19Q3 Kredit DLR 16 16 17 17 17 17 18 18 18 18 19 19 19 Non-cyclical capital requirements excl. Pillar II Note: As at 30 September 2019. Excess capital adequacy calcu- Pillar II add-on lated at the capital level (Common Equity Tier 1 capital, Countercyclical capital buffer, current Tier 1 capital or total own funds) with the shortest Conuntercyclical capital buffer, not yet in effect distance to the risk-based capital requirements for the institution. The risk-based capital requirements consist of the minimum capital requirement, the Pillar II add-on, Note: “Non-cyclical capital requirements excluding Pillar II” the SIFI buffer, the capital conservation buffer and the comprise fully phased-in requirements for minimum countercyclical capital buffer. It is assumed that the capital, capital conservation buffer and SIFI buffer. “Coun- exposure-weighted countercyclical capital buffer is 1.2 tercyclical capital buffer, not yet in effect” reflects buffer per cent for Danske Bank and 1.0 per cent for the other increases adopted by the Minister for Industry, Business institutions. and Financial Affairs, which will take effect in future Source: Danish Financial Supervisory Authority and own calcula- ­periods. tions. Source: Danish Financial Supervisory Authority and own calcula- tions.

in a few institutions, the distance to the combined II add-ons has increased in more institutions. Third- capital buffer requirement is less than 2 per cent of ly, it has been decided to raise the countercyclical risk-weighted exposures, cf. Chart 22. The excess capital buffer rate for credit exposures in Denmark capital adequacy of a few systemic institutions is from the current level of 1.0 per cent to 1.5 per cent thus low, not least in the light of expectations of as at 30 June 2020 and 2.0 per cent as at 30 Decem- higher capital requirements in the coming years, cf. ber 2020. Moreover, the Systemic Risk Council has below. announced that it expects to recommend a further increase of the buffer in the 1st quarter of 2020, un- Institutions should ensure an appropriate distance less the build-up of risk in the financial system slows between capital targets and capital requirements down considerably. All systemic institutions comply with their current capital targets. This also applies to institutions with A higher countercyclical capital buffer rate will not currently low excess capital adequacy. That is why make the institutions less resilient. However, given Danmarks Nationalbank finds that these institutions unchanged capitalisation, they will, all else equal, be should reconsider their capital targets to ensure an at a higher risk of becoming subject to restrictions appropriate distance between capitalisation and on payment of dividends and interest on Additional capital requirements. Tier 1 capital.

The need to reconsider capital targets is driven by The institutions can improve their capitalisation by several factors. Firstly, weaker earnings will – as men- issuing more equities or retaining a larger share of tioned previously – reduce the institutions’ first line profits. Disbursements to shareholders in the form of defence against losses. Secondly, the level of Pillar of dividends and buy-backs were historically high ANALYSIS — DANMARKS NATIONALBANK 16 FINANCIAL STABILITY – 2ND HALF 2019

in 2015-18. In 2019, the level of disbursements will apparently be considerably lower, cf. Chart 23. Lower Decline in disbursements Chart 23 disbursements contribute to increasing capitalisa- to shareholders tion, making it easier to meet higher capital targets. Kr. billion Bank Rescue Packages I and II 35

Solvency stress test emphasises 30 the need to reconsider capital targets 25 The need to reconsider capital targets is under- Profit/loss for the year 20 scored by the findings in Danmarks Nationalbank’s 15 most recent stress test of the banks’ capitalisation.6 It 10 shows that a few of the systemic banks will fall short of their capital buffer requirements in a severe stress 5 scenario – even if the countercyclical capital buffer is 0 assumed to be released. The buffer can be released -5 00 02 04 06 08 10 12 14 16 18 with immediate effect if the financial system becomes Dividend Share buy-backs stressed. But in some scenarios, one or more institu- tions could encounter difficulties without release of Note.: Systemic banks. The banks’ scope for distributing divi- the buffer. dends was limited under Bank Rescue Packages 1 and 2. Buy-backs and dividends in 2019 cover the period January-September. The non-systemic banks’ accumulation of capital Source: Danmarks Nationalbank. also seems to have declined. Several of the small non-systemic banks also breach their capital buffer requirements in Danmarks Nationalbank’s stress test. Box 2 How­ever, these banks’ breaches of their buffer re- The December 2017 Basel reform quirements are not found to pose a threat to financial stability. In December 2017, the Basel Committee on Banking Supervision, BCBS, completed a major reform package, issuing revised standards for credit risk, operational risk Denmark is one of the EU member states and counterparty risk (CVA risk) and introducing a new where the new Basel rules are expected floor, i.e. the output floor, under the institutions’ risk to have the strongest impact weights. The reform package is supplemented by a new The forthcoming implementation of the December set of rules on market risk (Fundamental Review of the 2017 Basel reform, cf. Box 2, is also a reason for Trading Book), issued in a new version in January 2019. the systemic institutions to reconsider their capital planning. In August 2019, the European Banking The BCBS has set 1 January 2022 as the deadline for timely implementation of the new standards, but with the Authority, EBA, published an analysis of the impacts possibility of gradual phasing in of the output floor by 1 of the reform, which shows that Danish institutions January 2027. – together with institutions in other Northern Euro- pean countries – are among those that can expect In October 2019, the European Commission launched a the largest increases in capital requirements.7 A public consultation on European implementation of the reform package with 3 January 2020 as the deadline for large share of the increases is attributable to the replies. The Commission is expected subsequently to introduction of the output floor. The output floor publish an overall proposal for implementation in the EU. means that the institutions’ risk-weighted exposures cannot constitute less than 72.5 per cent of the As regards the new rules on market risk, it was decided, risk-weighting calculated using the standardised when the revised Capital Requirements Regulation, CRR 2, was adopted in May 2019, that these rules should ini- tially be applied as a reporting requirement in the EU.

6 For more information about the stress test, see Danmarks Nation- albank, Banks are less resilient to stress, Danmarks Nationalbank Analysis (Stress test), No. 24, November 2019.

7 Cf. European Banking Authority, EBA, Basel III Reforms: Impact Study and Key Recommendations, August 2019 (link). ANALYSIS — DANMARKS NATIONALBANK 17 FINANCIAL STABILITY – 2ND HALF 2019

The largest Danish credit institutions are among those with the highest capitalisation Chart 24 measured in relation to risk-weighted exposures

Common Equity Tier 1 ratio Leverage ratio

CET 1 capital in per cent of risk-weighted exposures Tier 1 capital in per cent of unweighted exposures 25 10 Simple average Simple average New minimum 20 8 requirement

15 6

10 4

5 2

0 0 UK UK Italy Italy Spain Spain France Austria Ireland Finland Norway France Sweden Ireland Austria Belgium Finland Norway Sweden Germany Denmark Belgium Germany Denmark Netherlands Netherlands

Note: As at 30 June 2019. The comparison is based on data from the institutions that participated in the EBA’s most recent stress test from 2018, although a few countries/outliers have been excluded. The minimum requirement of the leverage ratio will become effective in mid-2021. Source: SNL Financial and interim reports.

approaches, even though IRB approaches based on capital targets are used – is that the largest Danish loss experience warrant lower risk weighting.8 institutions have a large share of relatively safe loans secured on real estate. The use of IRB models for The institutions’ capital ratios are high according calculation of risk weights results in similarly low to risk-based measures due to low risk weights risk-weighted exposures for the institutions. This also The international credit assessment agencies find explains why Denmark is one of the countries where that the systemic credit institutions are well capital- the output floor is expected to bring the strongest ised. By European comparison, the largest Danish increases in capital requirements. institutions are among those with the highest cap- italisation when the comparison is based on risk- MREL compliance reached as at 1 July 2019 based measures, cf. Chart 24 (left). A non-risk-based The institutions’ capital requirements are supple- approach using the leverage ratio shows that these mented by a minimum requirement for own funds institutions’ capitalisation is either lower than or and eligible liabilities, MREL. The MREL is to ensure close to a European average, cf. Chart 24 (right).9 that an institution can be resolved under the chosen resolution strategy if it becomes failing or likely to The reason for the very different results in relation fail. The general resolution principle for systemic to the average – depending on whether risk-based institutions is that it should be restructured and sent

8 According to the EBA analysis, the Tier 1 capital requirements of Dan- ish banks are expected to increase by around 40 per cent. A Danish expert group established by the Minister for Industry, Business and Financial Affairs previously estimated the effect for the systemic credit institutions at an increase in capital requirements of 34 per cent on average. The difference between the two estimates can be attributed to such factors as different assumptions and populations.

9 The leverage ratio is calculated as Tier 1 capital as a percentage of unweighted exposures. ANALYSIS — DANMARKS NATIONALBANK 18 FINANCIAL STABILITY – 2ND HALF 2019

back to the market with sufficient capitalisation to ensure market confidence. Mortgage lending to households Chart 25 is increasing, while bank lending All systemic institutions reached MREL compliance as is continuing its decline at 1 July 2019, when the requirement was to be met for the first time.10 Excess MREL relative to the new Kr. billion Kr. billion 400 4,000 requirement is limited, however, and results from Large banks Danmarks Nationalbank’s stress test show that the 350 3,500 systemic institutions will need to issue substantial 300 3,000 Milliarder 250 2,500 amounts of MREL-eligible instruments in a stress Mortgage credit institutions scenario. Danmarks Nationalbank expects the institu- 200 (right-hand axis) 2,000 tions to continue building up MREL-eligible instru- 150 1,500 ments in the coming years. 100 1,000 Medium-sized banks The MREL can be met, inter alia, by issuing non-pre- 50 500 ferred senior debt. Typical issuances by Danish 0 0 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 institutions have maturities of 3-5 years. Maturities for these issuances are generally longer in the rest of the EU, typically 5-10 years. Given that periods of Note: Lending to Danish households by large and medi- um-sized banks and mortgage credit institutions, no or limited capital market access may arise, the 3-month moving averages, nominal values. The most institutions should focus on the length and maturity recent observation is from end-September 2019. Source: Danmarks Nationalbank. profiles of their issuances to avoid strong concentra- tion of maturing issuances and breach of the MREL.

properties and other real estate categories account Homeowners opt for fixed rate for 12 and 4 per cent, respectively. Lending is on the mortgage loans decrease for these categories.

Bank lending is declining overall. But the medium­ Falling bank lending to households sized banks are still taking market shares from the and increased mortgage lending large banks despite the receding lending growth Lending to households by large and medium-sized over the past year, although to a lesser extent than banks and mortgage credit institutions has risen by previously. Lending growth for the medium-sized 1.2 per cent on average over the past year, cf. Chart banks is 2.8 per cent, while it is declining by 6.3 per 25. Lending growth is driven almost exclusively by cent for the large banks. mortgage credit, showing a growth rate of 2.2 per cent. The growth in mortgage lending is driven Growth in mortgage lending is almost by loans secured on owner-occupied housing and exclusively driven by fixed rate loans summer cottages, showing a growth rate of 3.1 per The increase in mortgage lending for owner-occu­ cent. Loans secured on owner-occupied housing and pied housing and summer cottages is driven by summer cottages make up 84 per cent of mortgage growth in fixed rate loans. Fixed rate mortgage loans lending to households. Loans secured on agricultural have grown by 9.3 per cent on average over the past

10 Spar Nord received SIFI status in January 2019, but is not obliged to observe the new MREL in full until the beginning of 2022. Mortgage credit institutions are exempt from the MREL, but must instead observe a debt buffer requirement of 2 per cent of the institution’s total lending. As from January 2022, groups comprising banking and mortgage credit activities are subject to a combined minimum requirement, under which the sum of capital, MREL and debt buffer requirements must be at least 8 per cent of the group’s balance sheet. ANALYSIS — DANMARKS NATIONALBANK 19 FINANCIAL STABILITY – 2ND HALF 2019

year, and growth has accelerated over the period, cf. Chart 26. Variable rate loans have decreased by 1.3 The increase in mortgage lending for Chart 26 per cent on average. Fixed rate loans now account owner-occupied housing and summer for 46 per cent of mortgage lending, against 33 cottages is driven by fixed rate loans per cent at the beginning of 2014. This means that Growth contribution, percentage points households have become more resilient to interest 7 rate changes related to their debt. 6 Fixed rate 5 Growth, year-on-year The falling interest rates in 2019 have made it more 4 attractive for many homeowners to refinance their 3 2 mortgage loans. During the current wave of refi- 1 nancing, a record-high number of homeowners have 0 chosen to remortgage. During the first 9 months of -1 Variable rate the year, prepayments have amounted to no less -2 than kr. 300 billion for household mortgage loans -3 secured on owner-occupied housing and summer -4 2015 2016 2017 2018 2019 cottages. Besides cases of refinancing, prepayments also include termination of loans in connection with property sales or for other reasons. Note: Mortgage lending to households for owner-occupied housing and summer cottages, nominal values. The most recent observation is from September 2019. Some homeowners choose to raise larger loans Source: Danmarks Nationalbank. when refinancing. In previous waves of refinancing, households used the additional mortgage credit for consumption and home improvements and for Fixed rate loans are increasing most Chart 27 11 reducing other debt. strongly around the large cities

Strongest lending growth in fixed rate mortgage loans around the large cities Fixed rate mortgage lending is increasing in all municipalities, cf. Chart 27. Growth is strongest in Growth, Copenhagen and Copenhagen environs and around per cent 15 the large cities.

10

Variable rate mortgage lending is decreasing in 5 most municipalities. However, some municipalities in Copenhagen and Copenhagen environs stand out in that variable rate mortgage debt is also increas- ing, albeit at a considerably lower growth rate than for fixed rate loans. Variable rate mortgage debt is falling in all other parts of Denmark.

The institutions report reduced risk appetite for households The institutions report almost unchanged credit standards for households in the 3rd quarter of 2019, while the competitive pressure has intensified a little. According to Danmarks Nationalbank’s lending Note: Average annual growth in fixed rate mortgage lending to households, owner-occupied housing and summer cottages, nominal values. The most recent observation is from September 2019. Source: Danmarks Nationalbank.

11 Cf. Danmarks Nationalbank, Mortgage refinancing supports private consumption, Danmarks Nationalbank Analysis, No. 17, September 2019. ANALYSIS — DANMARKS NATIONALBANK 20 FINANCIAL STABILITY – 2ND HALF 2019

survey, the demand for mortgage loans from existing customers has soared, cf. Chart 28. The mortgage Substantial demand for mortgage Chart 28 credit institutions also report rising demand from loans from existing household custom- new customers. According to some of the institu- ers during the wave of refinancing tions, some homeowners raise additional mortgage Net balance loans when refinancing, which are partially used to 60 reduce bank debt. Increased demand 40

Both banks and mortgage credit institutions have 20 reported reduced risk appetite in recent quarters, cf. 0 Chart 29. Especially the large banks and mortgage credit institutions are reporting reduced risk appe- -20 tite. Some institutions state that the reduced risk -40 appetite is due to the changes to good practice for Reduced demand -60 lending for housing purposes and the guidelines on 2015 2016 2017 2018 2019 prudent credit assessment when granting housing Banks Mortgage credit institutions loans in growth areas. Moreover, some institutions report that credit approval will to some extent be Note: Lending survey for households. The net balance may lie granted centrally in future. within the interval -100 to 100. A positive (negative) net balance means that credit managers of the institutions in question have, overall, i.e. lending-weighted, stated that demand from existing customers has increased (decreased) relative to the preceding quarter. The most Strong competition recent observation is from the 3rd quarter of 2019. for corporate customers Source: Danmarks Nationalbank.

Growth in lending to corporate The institutions report reduced Chart 29 customers continues at a moderate pace risk appetite to households Lending to corporate customers by large and me­ Net balance dium-sized banks and mortgage credit institutions 30 has risen by just under 4 per cent over the past year, Increased risk appetite 20 cf. Chart 30. Lending growth has declined a little over the last half year, driven by the large banks’ 10 decline in average annual growth. On average, lend- 0 ing by the large banks has increased by 2.1 per cent over the past year, while the figure for medium-sized -10 banks is 4.5 per cent. The market shares of the large -20 and medium-sized banks are unchanged. Growth Reduced risk appetite -30 in lending to corporate customers is still driven by 2015 2016 2017 2018 2019 mortgage credit. The average lending growth for Banks Mortgage credit institutions mortgage credit institutions has been 4.4 per cent over the past year, and lending growth is showing a slight upward tendency. Note: Lending survey for households. The net balance may lie within the interval -100 to 100. A positive (negative) net balance means that credit managers of the institutions in Intensified competitive pressure question have, overall, i.e. lending-weighted, stated that they have increased (reduced) their risk appetite relative and falling interest margins for corporate lending to the preceding quarter. The most recent observation is According to Danmarks Nationalbank’s lending sur- from the 3rd quarter of 2019. Source: Danmarks Nationalbank. vey, credit standards for corporate customers were more or less unchanged in the 3rd quarter of 2019. Both banks and mortgage credit institutions still report that competitor behaviour has contributed to their easing of credit standards, cf. Chart 31. Several banks expect the competitive pressure to increase ANALYSIS — DANMARKS NATIONALBANK 21 FINANCIAL STABILITY – 2ND HALF 2019

Lending growth to corporate custom- Chart 30 Increasing competitor pressure Chart 31 ers is still driven by mortgage credit for corporate customers

Kr. billion Net balance 800 50 Increased competition 40 700 30 600 20 Mortgage credit institutions 500 10 0 400 -10 Large banks 300 -20 -30 200 Reduced competition -40 Medium-sized banks 100 -50 2015 2016 2017 2018 2019 0 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 Banks Mortgage credit institutions

Note: Lending to Danish non-financial corporations by Note: Lending survey for corporate customers. The net balance large and medium-sized banks and mortgage credit may lie within the –100 to 100 interval. A positive (nega­ institutions. 3-month moving averages. The most recent tive) net balance means that credit managers of the observation is from September 2019. institutions in question have, overall, i.e. lending-weight- Source: Danmarks Nationalbank. ed, stated that competitor behaviour has contributed to easing (tightening) of credit standards relative to the preceding quarter. The most recent observation is from the 3rd quarter of 2019. Source: Danmarks Nationalbank. further in the 4th quarter, while the mortgage credit institutions expect the competitive pressure to re- main unchanged. Competition for corporate customers Chart 32 puts pressure on margins

The strong competitive pressure is primarily reflected Per cent in the banks’ interest margins. The large and medium­ 4.5 sized banks’ interest margins continue to decline, Medium-sized banks cf. Chart 32. The interest margin is defined as the 4.0 difference between the banks’ average lending and 3.5 deposit rates for corporate customers. The credit 3.0 quality can help to explain why the medium-sized Total banks have higher interest margins. The share of 2.5 corporate lending with normal credit quality is 55 per Large banks cent for the medium-sized banks and has improved 2.0 slightly over the last half year. But the credit quality is 1.5 still considerably lower than that of the large banks, 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 whose share of corporate lending with normal credit quality is 86 per cent. It is important for the institu- Note: Interest margins on outstanding business, excluding repo tions to maintain a sound credit quality and to ensure transactions, 3-month averages. The interest margin is that the inter­est margin reflects the risk on the loan. calculated as lending rates less deposit rates. The spread between large and medium-sized interest margins Since losses do not materialise until the economy re- reflects differences in the loans’ credit risks, but also verses, the institutions should assess firms’ resilience other factors such as product structures. The most recent throughout the business cycle. observation is from September 2019. Source: Danmarks Nationalbank.

Demand for mortgage loans from corporate customers is increasing, while it is stagnating for bank loans Corporate customers’ demand for mortgage loans is increasing. Over the last two quarters, mortgage ANALYSIS — DANMARKS NATIONALBANK 22 FINANCIAL STABILITY – 2ND HALF 2019

credit institutions have reported rising demand from new customers, cf. Chart 33. The demand from exist- Increasing demand for mortgage Chart 33 ing customers is also reported to have grown. De- loans and falling demand for bank mand has increased from large, small and medium­- loans from new corporate customers sized corporate customers. According to some of the Net balance mortgage credit institutions, refinancing activity is 50 Increased demand high. This could help to explain the last two quarters’ 40 intensified competitive pressure among mortgage 30 credit institutions. 20 10 0 Both large and medium-sized banks are report- -10 ing lower demand from new customers. The lower -20 -30 demand contrasts with the situation in recent years Reduced demand -40 when medium-sized banks reported increasing de- -50 mand for loans from new customers. The stagnating 2015 2016 2017 2018 2019 demand could contribute to even tougher competi- Banks Mortgage credit institutions tion.

A good funding climate for Danish firms Note: Lending survey for corporate customers. The net balance may lie within the interval -100 to 100. A positive (nega- There is still a good funding climate for firms. Few tive) net balance means that credit managers of the in- firms are reporting restrictions on the access to stitutions in question have, overall, i.e. lending-weighted, stated that demand from new customers has increased funding, cf. Chart 34. According to analyses from the (decreased) relative to the preceding quarter. The most Confederation of Danish Industry12, 65 per cent of recent observation is from the 3rd quarter of 2019. Source: Danmarks Nationalbank. firms find that funding access is good or very good, which is a historically high level. Firms’ funding ac- cess depends on their key ratios, meaning that firms with higher returns on assets and solvency ratios Few firms report Chart 34 find that their funding access is better. restricted funding access

Share, per cent 25 Liquidity is robust, but sensitive to deposit withdrawal 20

15 Services Short-term LCR observed with a certain margin 10 Danish banks observe the short-term Liquidity Building and construction Coverage Ratio, LCR, requirement, which is to ensure 5 that the banks have sufficient liquid funds to with- Manufacturing stand a 30-day severe liquidity stress scenario, cf. 0 Chart 35. In the current climate of favourable market 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 liquidity, it is natural for the banks to have a certain amount of excess liquidity to be able to withstand Note: Share of firms stating financial restrictions as the reason liquidity shocks in periods of less accessible liquidity. for production constraints. The most recent observation is from September 2019. Source: Statistics Denmark.

12 Confederation of Danish Industry, Historisk lave renter på lån til virksomhederne (Historically low interest rates on corporate loans – in Danish only), September 2019, and Confederation of Danish Industry, Historisk nemt at låne penge (Borrowing is historically easy – in Danish only), April 2019. ANALYSIS — DANMARKS NATIONALBANK 23 FINANCIAL STABILITY – 2ND HALF 2019

Pct. The banks comply with the short-term Liquidity Coverage Ratio, Chart 35

LCR,800 with a certain margin

Systemic700 banks Non-systemic banks

Per600 cent Per cent 250 800 500 700 200 400 600 150 500 300 400 100 200 300 Minimum requirement 200 50100 100 Minimum requirement 0 0 0 Jul 18 Jul 19 Jul 18 Jul 19 Jul 19 Jul 17 Jan 18 Jan 19 Jun 18 Jun 19 Oct 19 Oct Jan 18 Jan 19 Okt 18 Okt Jun 18 Jun 19 Apr 18 Apr 19 Jan 19 Jan 17 Jan 18 Oct 19 Oct Feb 18 Feb 19 Jun 19 Jun 17 Jun 18 Sep 18 Sep 19 Okt 18 Okt Dec 18 Apr 18 Apr 19 Oct 18 Oct 19 Oct 16 Oct 17 Oct Mar 18 Mar 19 Aug 18 Aug 19 Feb 18 Feb 19 Sep 18 Sep 19 Apr 19 Apr 17 Apr 18 Nov 18 Nov Dec 18 Feb 19 Feb 17 Feb 18 May 18 Sep 18 Sep 19 Sep 16 Sep 17 Mar 18 Mar 19 Aug 18 Aug 19 Dec 18 Dec 16 Dec 17 Nov 18 Nov Mar 19 Mar 17 Mar 18 Aug 19 Aug 17 May 18 Nov 18 Nov 16 Nov 17 Nov May 19 May 17 May 18 May 19 May May 19 May

Median 10th percentile 90th percentile

Note: The LCR (Liquidity Coverage Ratio), which must be higher than 100 per cent, is calculated as the bank’s liquid assets divided by net cash outflows over a 30-day stress period. The most recent observation is from end-October 2019. Source: Danmarks Nationalbank.

Besides the general LCR requirement, banks must increases, all else equal14, as the value of the liquid also ensure a sufficient match between net outflow assets may fall more than expected, e.g. if one of the and liquid assets in relevant currencies. In addition, counterparties encounter difficulties. the systemic Danish banks must observe a liquidity requirement in currencies making up more than 5 per Overall, the systemic banks’ liquidity distribution is cent of their liabilities. However, the requirement does relatively concentrated in terms of both asset classes not apply to Swedish kronor and Norwegian kroner. and counterparty exposure, cf. Chart 36. In the 3rd quarter of 2019, all institutions complied with the liquidity requirement in currencies. Covered bonds, mainly mortgage bonds, clearly ac- count for the largest share of liquidity. This reflects that The banks’ total liquidity stock mortgage bonds account for the largest share of the mainly consists of mortgage bonds market for highly liquid assets in Danish kroner. Alter- The banks’ liquidity should be diversified in order natively, the banks could assume either a higher credit to reduce concentration risks. Concentration may risk or currency risk, which does not seem appropriate. occur for asset classes as well as counterparties. The LCR partially takes concentration into account The liquidity concentration reflects concerning asset classes, but not counterparties.13 a limited number of issuers If liquidity is highly concentrated on one asset class All mortgage bonds in the liquidity stock have high or a few counterparties, the institution’s liquidity risk credit ratings. Despite the low credit risk on the

13 In the LCR, liquid assets are denoted high quality liquid assets, HQLA. 14 The risk is affected by the type of counterparty/counterparties. For They are a subset of the banks’ total liquidity stock. When calculating example, an exposure solely to Danmarks Nationalbank will not be HQLA, at least 30 per cent must be government-guaranteed assets or regarded as entailing elevated liquidity risk. central bank deposits. In addition, there are requirements as to the maximum shares of various asset classes. The most important factor for Danish institutions is that mortgage bonds must not exceed 70 per cent. ANALYSIS — DANMARKS NATIONALBANK 24 FINANCIAL STABILITY – 2ND HALF 2019

individual issuers, the institution will still assume an elevated liquidity risk if the liquidity stock is not The banks’ liquidity consists Chart 36 diversified on several issuers. mainly of assets issued by financial counterparties

Basically, a bank’s exposure to a counterparty should Kr. billion not be so concentrated that it may encounter liquid­ 900 ity difficulties if the counterparty runs into difficulties. 800 Other 700 The concentration on individual issuers is assessed to 600 Sovereigns Central be relatively limited in relation to a benchmark for di- 500 bank versification. The benchmark is based on the individ- 400 Bank Largest ual mortgage bond issuers’ market shares. Deviations 300 10 ex- posures MFI from such a benchmark show that the systemic banks 200 Mortgage credit hold more than their market shares would justify for 100 three out of the six issuers of mortgage bonds, cf. 0 Liquidity Top 10 MFI Chart 37. Conversely, the exposure to two of the is­ holding suers is lower than justified by their market shares.

Note: The liquidity stock consists of unencumbered assets that It should be noted that a 1:1 reflection of the bench- a bank can use for liquidity purposes, i.e. a broader de­ mark would not be appropriate. The reason is that finition that the LCR definition of liquid assets. The stock is compiled without haircuts. MFI stands for monetary such reflection would imply that institutions would financial institutions. The chart shows the systemic banks’ also have a large share of own issuances in their liquidity stock at end-August 2019. liquidity stock, since the systemic banks and mort- Source: Danmarks Nationalbank. gage bond issuers overlap. Holding own issuances in the liquidity stock entails a special risk, since own bonds can be expected to be difficult to sell in the The concentration of mortgage bonds Chart 37 event of uncertainty about the institution’s solvency. on individual counterparties varies

SIFIs' deviation from the volume of A sensitivity analysis of the banks’ mortgage bonds issued, percentage points 8 liquidity shows resilience to market stress ... More exposed Danmarks Nationalbank has developed a tool for 6 assessment of the banks’ liquidity position. The sen- 4 sitivity analysis provides information on the factors 2 affecting the banks’ liquidity position and how fast a 0 bank may run into liquidity difficulties. The sensitiv- -2 ity analysis thus supplements the statutory liquidity -4 Less exposed requirement, LCR, and the future NSFR (Net Stable -6 Funding Ratio). -8 -10 Insti- Insti- Insti- Insti- Insti- Insti- The sensitivity analysis is based on three scenarios: tution A tution B tution C tution D tution E tution F a market-specific scenario, an institution-specific scenario and a combination scenario, cf. Chart 38. Note: The chart shows the deviation of SIFIs’ average stock of The scenarios and assumptions of the liquidity mortgage bonds from the distribution of outstanding bonds. sensitivity analysis are explained in more detail in Source: Danmarks Nationalbank. Box 3.

The sensitivity analysis shows that the liquidity of the systemic banks is generally robust to a market downturn entailing that the banks cannot expect to be able to renew existing loans and issuances when they mature. There is, however, considerable variation between the banks. A situation where the banks are not able to renew existing loans is rem- ANALYSIS — DANMARKS NATIONALBANK 25 FINANCIAL STABILITY – 2ND HALF 2019

The institutions are more exposed Chart 38 The outflow rate for deposits Chart 39 to an institution-specific shock is decisive for how long the liquidity than a marked-specific one position remains positive

Per cent Per cent 100 100 Market-specific 80 80 Halving of outflow rate 60 60 40 Institution-specific 40 20 Institution-specific in the model 20 0 Combination -20 0 Double outflow rate -40 -20 1 week 1 1 month 1 week 2 months2 months3 months4 5 months months6 1 1 month 2 months2 months3 months4 5 months months6

Note: The chart shows the sensitivity to a change to the as- Note: The chart shows the systemic banks’ net liquidity over sumption of how fast demand deposits are withdrawn. time in various outflow scenarios. Net liquidity is defined Source: Danmarks Nationalbank. as the liquidity stock adjusted for net outflow over time. Data from end-August 2019. Source: Danmarks Nationalbank.

iniscent of what hit the financial markets after the risks affecting liquidity withdrawals in a crisis. They collapse of Lehman Brothers in September 2008. can obtain this understanding by, inter alia, actively using stress tests for liquidity and risk management … but the liquidity is purposes e.g. via frameworks for survival horizon, sensitive to deposit withdrawals outflow rates, etc. It is also important to perform The sensitivity analysis also shows that the system- robustness checks of the models’ sensitivity to the ic banks’ liquidity position is more sensitive to an individual parameters, e.g. by assuming different institution-specific shock, for instance with doubt outflow rates. about whether an individual bank will survive and with depositors consequently withdrawing deposits. New, more risk-sensitive liquidity The most important parameter for explaining the requirement for mortgage credit institutions difference is by how much and how fast household In 2015, the Danish Financial Supervisory Authority and corporate customers withdraw their deposits introduced a Pillar II liquidity requirement for mort- from the bank. The aim of the combination scenario gage credit institutions, i.e. a minimum requirement is to illustrate a situation of both risks materialising at of 2.5 per cent of lending, to cover risks not directly the same time. Although this is a very severe scenario, covered by the short-term LCR. An example is the the sensitivity analysis shows that the banks still have risk associated with refinancing auctions. In Septem- time to react before they run into liquidity difficulties. ber 2019, the Danish Financial Supervisory Author- ity revised the model to take into account a new Even small changes to the assumption of how fast requirement for reserving liquid assets equivalent to customers withdraw their deposits may have a con- 2 per cent of lending in mortgage credit institutions. siderable impact on when a bank’s liquidity runs dry, This requirement is an element of the forthcoming cf. Chart 39. For instance, a bank’s survival horizon EU Directive on regulation of mortgage bonds and will be reduced by almost three months if the out- other covered bonds. The revised model is more flow rate doubles. risk-sensitive than the existing one, as the future liquidity requirement will fluctuate in step with, inter It is important that the banks themselves have a alia, the size of refinancing auctions and the volume deep understanding of the sensitivity to various of open refinancing where the funding behind the ANALYSIS — DANMARKS NATIONALBANK 26 FINANCIAL STABILITY – 2ND HALF 2019

Explanation of the scenarios and assumptions in the liquidity sensitivity analysis Box 3

The projection of the liquidity position is based on calcula- transactions, e.g. household customers are not withdrawing tions of the banks’ excess liquidity under various assump- demand deposits. Relatively high haircuts are assumed for tions of outflow rates regarding both contractual flows and the liquidity stock, as it is expected to be difficult to realise demand deposits as well as the realisation value of the the assets in a general banking sector stress situation. For various assets in the liquidity stock. example a 7 per cent haircut for mortgage bonds are as- sumed, matching the haircut in the short-term LCR. The scenarios have been constructed under the assumption that the liquidity stock is to be sold in the market. This gives Institution-specific scenario a picture of how long the bank is able to manage its own Simulates stress at individual bank level, e.g. due to rumours liquidity situation without resorting to Danmarks National- of difficulties. On the one hand, it is assumed that the bank bank’s facilities. loses access to the funding markets with simultaneous considerable losses of deposits. On the other hand, it is The outflow rates for demand deposits are inspired by the assumed that the lending balance is gradually reduced outflow weights in the ECB’s 2019 sensitivity analysis, but via prepayments from corporate customers, and that the have been adjusted to fit the different scenario construc- liquidity stock can be realised at relatively limited haircuts tions in the two sensitivity analyses. The outflow rates are (e.g. a haircut of 3 per cent is applied to mortgage bonds). found to be severe, but far from inconceivable in a historical However, assets issued by the institution are assumed to be context. The outflow is assumed to be most severe in the realised at a large loss of value of 50 per cent. first 30 days, followed by a gradual reduction. Combination scenario Given that the sensitivity analysis is constructed at institu- Simulates a general financial sector stress situation where tion level, it addresses the degree of banks’ liquidity without the institution is also hit by an institution-specific event. For the influence from affiliated mortgage credit institutions, if example rumours of imminent failure during a general crisis. applicable. This is a very difficult situation, in which the bank’s access to funding markets can be expected to be gone, simultan­ The analysis operates with three scenarios: eously with a deposit outflow from the balance sheet. But it •• Market-specific scenario is not possible to reduce lending, given that the whole sec- •• Institution-specific scenario tor is under pressure, and at the same time the realisation •• Combination scenario. value of the liquidity stock has been reduced considerably. Technically, the most severe parameters from the two above Market-specific scenario scenarios are applied. Simulates general stress in the banking sector with funding markets closing, but with no strong effect on daily customer

1. Information on and results of the ECB’s sensitivity analysis are available at the ECB website (link).

customers’ refinancing of mortgage loans has not yet been raised. The new requirement is thus expected to clarify the fluctuations in mortgage credit institu- tions’ liquidity positions resulting from large waves of refinancing when interest rates change. Danmarks Nationalbank takes a positive view on enhancing the risk sensitivity of the liquidity requirement. The new liquidity requirement must be reported as from December 2019 and is expected to be implemented no later than the European requirement for reserv- ing liquid assets equivalent to 2 per cent of lending become effective. ANALYSIS — DANMARKS NATIONALBANK 27 FINANCIAL STABILITY – 2ND HALF 2019

Appendix: Data for the analysis

Institutions in the analysis by balance sheet total as at 30 June 2019, kr. million Table 1

Systemic credit institutions Amount Non-systemic banks

Danske Bank (including Realkredit Danmark) 3,290,789 Arbejdernes Landsbank 57,884

Nykredit Realkredit (including Nykredit Bank) 1,540,708 Ringkjøbing Landbobank 52,426

Jyske Bank (including Jyske Realkredit) 646,416 Sparekassen Kronjylland 28,415

Nordea Kredit 455,230 Sparekassen Sjælland-Fyn A/S 24,885

DLR Kredit 165,730 Lån & Spar Bank 24,092

Sydbank 152,073 Sparekassen Vendsyssel 22,067

Spar Nord 89,354 Vestjysk Bank 21,592

Systemic groups, total 6,340,300 Jutlander Bank 19,120

Den Jyske Sparekasse 15,736

Systemic banks Alm. Brand Bank 13,913

Danske Bank 2,324,499 Non-systemic banks, total 280,131

Jyske Bank 304,701

Nykredit Bank 214,260 Mortgage credit institutions

Sydbank 154,372 Nykredit Realkredit (including Totalkredit) 1,472,231

Spar Nord 89,482 Realkredit Danmark 910,400

Systemic banks, total 3,087,314 Nordea Kredit 455,230

Jyske Realkredit 380,237

DLR Kredit 165,730

LR Realkredit 28,914

Mortgage credit institutions, total 3,412,741

Note: The balance sheet totals of systemic banks, non-systemic banks and mortgage credit institutions are stated at institution level, while the balance sheet totals of the systemic credit institutions are stated at group level with the exception of DLR Kredit and Nordea Kredit which are stated at institution level. Source: Danmarks Nationalbank.

The analysis applies the term “credit institutions” Supervisory Authority’s group 2, Saxo Bank has been when referring to the activities of both banks and omitted because of its business model. The grouping mortgage credit institutions. The term “bank” refers also applies back in time. specifically to entities carrying out banking activ­ ities. In the analysis and assessment of lending activity, focus is on the grouping of large and medium-sized The analysis of Danish credit institutions’ earnings, banks in Danmarks Nationalbank’s lending survey. liquidity and own funds comprises seven systemic Large banks are the Danish Financial Supervisory credit institutions. The analysis also includes the ­Authority’s group 1 plus the branch Nordea Danmark, non-systemic banks in the Danish Financial Super­ while medium-sized banks are the Danish Financial visory Authority’s group 2 in 2019. These institutions Supervisory Authority’s group 2 plus the branches are listed in Table 1. Unlike in the Danish Financial Handelsbanken and Santander Consumer Bank. ANALYSIS — DANMARKS NATIONALBANK FINANCIAL STABILITY – 2ND HALF 2019

ABOUT As a consequence of Danmarks National- Analyses are published continuously and ANALYSIS bank’s role in society we conduct analyses include e.g. assessments of the current of economic and financial conditions. cyclical position and the financial stability.

The analysis consists of a Danish and an English version. In case of doubt regarding the correctness of the translation the Danish version is considered to be binding.

DANMARKS NATIONALBANK HAVNEGADE 5 DK-1093 COPENHAGEN K WWW.NATIONALBANKEN.DK

This edition closed for contributions on 26 November 2019