ESRB European Systemic Risk Board : Notification by the Magyar Nemzeti Bank of Hungary on systemic risk buffer

Template for notifying the intended use of a systemic risk buffer (SRB)

Please send this template to

• •[e-mail] when notifying the ESRB;

• •[e-mail] when notifying the ECB;

• •[e-mail] when notifying the EBA.

Emailing this template to the above-mentioned addresses constitutes an official notification, no further official letter is required. In order to facilitate the work of the notified authorities, please send the notification template in a format that allows electronically copying the information.

1. Notifying national authority and scope of the notification
1.1 Name of the notifying authority	Magyar Nemzeti Bank (MNB)
1.2 Type of measure intended (also for reviews of existing measures)	Revise the SyRB rates
2. Description of the notified measure
2.1 Institutions covered by the intended SRB	The measure is to be applicable to all credit institutions authorised in Hungary (thereinafter: credit institutions) that are within the scope of the EU/575/2013 Regulation (CRR). The SyRB is applicable to all credit institutions with problem project exposures above the de minimis limit. Credit institutions shall apply the measure on a sub-consolidated basis. SyRB formed in CET1 capital will be required only on domestic exposures (RWA), thus the SyRB requirement excludes exposures to other Member States or to third countries. Institutions required to maintain other than zero SyRB rate were first identified in the second quarter of 2017 based on data provided for 31 March 2017. The SyRB rates of banks is revised annually, first in the second quarter of 2018 based on data provided for 31 March 2018. The MNB used its dedicated quarterly data reporting requirement to support a full review of the risks related to problem project exposures (including project loans and repossessed commercial real estate exposures) on a sub-consolidated level and to ensure the appropriate basis for quantifying the SyRB requirement.
2.2 Buffer rate (Article 133(11)(f) of the CRD)	As the MNB sets SyRB rates for this dedicated systemic risk up to 2 percent, the measure is subject to the 'notification only procedure' under Article 133 (11) of the 2013/36/EU Directive (CRD). Institution-specific SyRB rates are set in the range of 0 to 2 per cent. The size of the SyRB rate depends on the contribution of institutions to the systemic risk stemming from problem commercial real estate (CRE) exposures. Problem CRE exposures (hereinafter Problem CRE exposures) are defined for the purposes of the SyRB as: • the gross amount of domestic commercial real estate project loans if these: o are non-performing for more than 90 days, o are restructured, with the exception of loans restructured more

Date of template version: 2016-03-01

than a year ago that have become performing loans since, and o other project loans that are deemed non-performing by the financial institution; • the gross amount of domestic on-balance sheet held-for-sale commercial real estate. The contribution to the systemic risk is measured by the ratio of Problem CRE exposures to the sub-consolidated domestic Pillar I. capital requirement. The ratio of problem exposures to domestic Pillar I. capital requirements 0.00 - 29.99 % 30.00 - 59.99 % 60.00 - 89.99 % above 90.00 % Buffer rate +0.0% +1.0% +1.5% +2.0% A de minimis rule applies: institutions with problem exposures below HUF 5 billion are exempted from the SyRB requirement. Based on 31 March 2018 data, only one bank - CIB Bank Zrt. - is above the 30 percent threshold having problem exposures exceeding the HUF 5 billion de minimis limit. Thus, CIB Bank Zrt. will be obliged to maintain a SyRB of 1 percent to the domestic RWA from 1 July 2018 compared to the 2-percent SyRB rate effective from 1 July 2017 until 30 June 2018. The Raiffeisen Bank Zrt. will be obliged to maintain an SyRB of 0 percent to the domestic RWA from 1 July 2018 compared to a 1.5-percent SyRB rate effective from 1 July 2017 to 30 June 2018. SyRB rate to the domestic RWA 1 July 2017-30 1 July 2018-30 June 2018 June 2019 CIB Bank Zrt Raiffeisen Bank Zrt. +2.0% +1.5% +1.0% 0.0%
2.3 Exposures covered by the SRB	Only domestic exposures
3. Timing of the measure
3.1 Timing of the Decision	19 June 2018
3.2 Timing of the Publication	Banks were notified on 25 June 2018 and the press release is to be published on 3 July 2018.
3.3 Disclosure	All relevant information was communicated via a press release and the official website of the MNB. See:http://www.mnb.hu/en/financial-stability/macroprudential-policy/the-macroprudential-toolkit/instruments-to-limit-excessive-exposure-concentrations In the case of the current revision, the MNB has published a press release: • with the justification of the current actions, • with the names and applicable SyRB rates of the two institutions concerned.
3.4 Timing of Application	1 July 2018
3.5 Phasing in	The current notification is about the revision of the currently effective SyRB requirement from 1 July 2018.
3.6 Review/deactivation of the measure	The General Decision on the SyRB and the institution-specific requirements setting individual buffer rates will be reviewed at least annually. All decisions will be based on the ratio of domestic Problem CRE exposures to the sub-consolidated domestic Pillar I. capital requirement
4. Reasons for the intended SRB
4.1 Description of the long-term non-cyclical systemic risk in your Member State (Article 133(11)a of the CRD)	The persistently high ratio of the problem project exposures in the Hungarian banking sector was identified as a key macroprudential risk in 2014. Although non-performing portfolios could be regarded as a natural accompanying feature of banking, if they increase too fast and persist for too long, they pose severe problems to financial stability through several channels, finally hampering the banking sector's contribution to economic growth (e.g., in corporate lending). The NPL ratio of CRE project loans in the Hungarian banking sector was much higher (~25 percent) than the decreasing NPL rate of total corporate loans (~15 percent), and the rate was permanently stabilised on these high levels for more than 3 years in 2014. These facts, however, did not grasp the full range of the problem, as the stock of continuously restructured loans was also persistently high, and repossessed collateral from defaulted loans contributed also to the systemic risk. The stock of Problem CRE exposures therefore represented an issue even greater than NPL rates would suggest. The stock in question at the end of the second quarter of 2015 reached around HUF 730 billion (~EUR 2.4 billion) and made up approximately 45 per cent of total domestic project loans. Moreover, the large stock of Problem CRE exposures further increased stability risks due to being highly concentrated. By the announcement of the intended use of the SyRB at the end of 2014, 10 banks owned most of the stock of problem exposures, and 3 banks owned almost 70 per cent of all Problem CRE exposures. The exposure in question was not only nominally large; the stock could reach up to 1.5-3 times the size of the Pillar I. capital requirement of the affected institutions. Overall it was of public interest from a macroprudential point of view to intervene in this problematic segment and encourage institutions to either take steps in order to curb the volume of distressed portfolio exposures and/or to enhance their shock absorbing capacity to the level where systemic risks are also internalized.
4.2 Reasons why the dimension of the long-term non-cyclical systemic risk threatens the stability of the financial system in your Member State (Article 133(11)(b) of the CRD)	The high and persistent Problem CRE project loan NPL rate carried significant risk to financial institutions, as a large portion of the projects in question were backed by CRE that was not expected to become income (cash-flow) producing. Problem project exposures threatened financial stability through a number of channels: • Impact on profitability: As the stock of distressed CRE project loans increased, so did impairment losses, which negatively affected the profitability of banks, and which weakened the ability of institutions to accumulate capital. Furthermore, NPL stocks in such volume could tie down significant resources, which could again lead to deteriorating profitability (if, e.g., employees are assigned to workout activities instead of loan origination). The need to continuously finance the unimpaired part of non-performing loans also decreased profitability. The cost of financing could also increase due to high NPL rates, as

investors payed close attention to NPL in their risk evaluation; high NPL rates therefore could increase the risk premium on bank liabilities. • Impact on lending: As non-performing project loans tied down financing resources, the banking sector's lending capacity could weaken in case of effective liquidity limits, leading to decreased lending. High NPL rates could limit lending not only through lower capacity, but also through negatively affecting the willingness for lending: high NPL rates could make banks over-cautious, lowering their willingness to take on more risk in financing investment and growth in the corporate segment. • Impact on collaterals: A change in the value of collaterals or, in case of FX loans, in the exchange rate, could increase impairment losses. In an extreme scenario, in case of a deeper crisis, the value of collaterals could nosedive, leading to significant system-wide consequences in such moderately liquid markets like the one for commercial real estate. It is important to mention that the impact of Problem CRE project exposures was multiplicative in nature, i.e., the impact of the total system-wide risk was greater than the sum of the impacts of the risks of individual institutions due to, e.g., possible system-wide fire sales and the earlier moderately liquid market for the repossessed CRE.
4.3 Indicators used for the activation of the measure	The following indicators were used for the activation of the SyRB and the monitoring of the targeted macroprudential risk: • Total domestic problem project exposures over domestic Pillar I. capital requirement (calibration indicator) • Stock of total and domestic non-performing project loans and its ratio over total and domestic project loans • Stock of total and domestic restructured project loans and its ratio over total and domestic project loans • Stock of total and domestic problem project exposures and its ratio over total and domestic project exposures • Concentration of problem project exposures in the banking sector In case of all indicators, persistency through time was also taken into account.
4.4 Effectiveness and proportionality of the measure (Article 133(11)(c) of the CRD)	The SyRB requirement was a suitable instrument for the treatment of Problem CRE exposure risk as it was supposed to target a non-cyclical systemic risk in an EU Member State. Problem CRE exposures posed a structural systemic risk that was concentrated and had remained high for a relatively long time despite sequential Pillar 2 measures and posed a burden on the balance sheet of several major banks in Hungary. The SyRB requirement is supposed to target this risk as it can be levied on the combination of exposure classes that are deemed to be the most problematic. This measure is effective, as it increases the affected credit institutions' shock absorbing capacity to the level that internalizes systemic level risks. As a consequence, they are able to withstand the effects of collective exposure and collective inaction in problem resolution. On the other hand, due to the long phase-in period it was feasible for banks to wind down Problem CRE exposures, facilitating the mitigation of the identified systemic risk. The chosen definition of Problem CRE exposures (as explained above), the level of application and the targeted data reporting requirement minimised the room for the circumvention of the macroprudential measure. The measure is proportionate as it is directly linked to the contribution of the institutions to the systemic risk identified. It is derived from the ratio of the gross amount of Problem CRE exposures to the domestic Pillar I

capital requirement. The capital surcharge is levied on domestic RWA and was capped at 2 per cent. The "de minimis" limit of HUF 5 billion ensures that only systemically relevant exposures are taken into account.
4.5 Justification of inadequacy of existing measures in the CRD or in the CRR, excluding Articles 458 and 459 of the CRR, to address the identified risks (Article 133(11)(e) of the CRD)	The SyRB requirement is only applied to credit institutions that significantly contribute to the systemic risk of Problem CRE exposures. The SyRB requirement is aimed at strengthening their loss absorbing capacity to the level where systemic risks are also internalized. These systemic risks that come on top of institution-specific risks are complementary to those risks that have already been covered by additional Pillar 2 capital requirements. Due to the high geographical concentration of these properties and land in distress, a lot of non-performing CRE project loans have no prospect of becoming income generating. This concentration can bear systemic consequences. Capital buffers over Pillar 2 requirements are also warranted for scenarios of sudden exchange rate deterioration as the majority of these loans are denominated in FX and any simultaneous increase in the payment obligations can result in further fire sales or any other detrimental system-wide reaction. On the other hand, if banks react to the increased capital requirements by cleaning off Problem CRE portfolios, it can result in the mitigation of systemic risk, which can be followed by a decrease in the required SyRB level, calibrated during the yearly revision exercise. As a relatively long phase-in period was set (until the first quarter of 2017), affected institutions had the opportunity to accommodate and therefore to moderate or avoid the expected capital add-on levels. Other regulatory measures had proven to be insufficient for the mitigation of the systemic risk in question. Non-performing project exposures (both loans and repossessed real estates) were deemed significantly risky during the SREP for all relevant institutions in the last 3 years prior to 2014. The buffer levels were set at strictly high levels (the risk weights were increased by 100 per cent for both exposure types) during those three years, but the SREP capital add-on was not able to completely address the systemic risk of the exposures in question. Furthermore, the total SREP add-on is capped at 250% in Hungary, which means that there is a limit on the amount of capital add-on that can be prescribed for problem project exposures within the framework of the SREP. Therefore, the SREP capital add-on was not able to cover the full spectrum of the risks in question, and it was also limited in its capacity to ensure the necessary level of shock absorbing capacity or the necessary incentives for any prospective resolution of the problem, e.g. via asset cleaning. Article 124 and 164 measures of the Capital Requirements Regulation (CRR) are primarily devoted to preventing the emergence of credit/asset bubbles, and they do not differentiate among newly issued and outstanding exposures, thus they are not capable of tackling current systemic risks originated from problem project exposures. Moreover, in the case of project loan exposures, the value of the property materially depends upon the credit quality of the borrower, the exposures are deemed unsecured for the purposes of determining their risk weight. This means that there is no room for macroprudential measures within these articles to handle the targeted challenge. The case is similar for Article 458 measures applicable to contain risks stemming from asset bubbles in the CRE sector.
5. Cross-border and cross-sector impact of the measure
5.1 Assessment of cross-border effects and the likely impact on the internal market	As only Hungarian exposures are taken into account, the MNB does not expect significant impact of the measure on the functioning of the internal market. No significant negative spillover effects on other countries have been anticipated, as the buffer was prescribed based on domestic