Thank you for the opportunity to comment on the European Commission’s consultation on the review of the EU macro-prudential policy framework. We would like to share with you the following reflections that we hope will be taken into account by the Commission.
Question 1: Do you consider the degree of coordination between the different authorities in the current framework (i.e. ESRB, national macro-prudential authorities, Commission, Council, etc.) appropriate?
ESBG believes that there is room for improvement in the cooperation between the authorities. The authorities focus on their own targets and approaches and take into account the overlaps only very limitedly. As a result, it happens that the same issue in a bank is covered by different macro- and micro-prudential tools implemented by different authorities (or by the same authority in its different roles) in a different manner.
Furthermore, we would also like to emphasise the clear cut between the uses of the different tools according to their original purposes. If a tool is designed for a certain purpose, and the existing regulation is quite clear in this aspect, it should be used for that purpose only. For example: if the assessment of a bank in the SREP shows that in all assessed aspects the bank is broadly compliant with the requirements and implemented proper and effective processes, methods and systems to measure and cover the risks, there should not be a Pillar 2 capital add-on defined for that bank (and certainly not in the same percentage as for another bank, where serious deficiencies were found). Another example is the real estate portfolio, mentioned extensively in the Commission’s consultation paper.
ESBG also believes that the use of the macro-prudential tools is not always transparent and predictable. ESBG members who are active in different countries – SSM-countries, non-SSM but EU countries as well as non-EU countries – beyond inconsistencies between the ESRB’s, ECB’s and NCAs’ usage of the toolkit, have also observed frequent and unpredictable changes in the actual use of the toolkit. Under these circumstances, it is very hard to plan the capital position and create long-term solutions for the management and coverage of risks. Therefore, we would appreciate a clear and consistent legislation, leading to a similar implementation.
Question 2: Would you consider appropriate to expand the macro-prudential framework beyond banking? If deemed appropriate, what kind of systemic risks should be targeted and how?
In the interests of a level playing field, we advocate including the entire financial industry, e.g. insurance companies, funds and shadow banking entities, in the focus of macro-prudential instruments. In ESBG’s opinion, especially shadow banking should be targeted by developing suitable macro-prudential tools.
At present, however, it seems that banks are easiest to target in macro-prudential policy due to the broad availability of data. Yet they are not the only source of systemic risk. There needs to be a general discussion on whether the institutional approach is appropriate in the long term for reducing systemic risk or whether individual processes and the associated products would not be more suitable as the subject of macro-prudential policy.
Question 3: Do you see a need to strengthen the coordination between designated and competent authorities when using stricter Pillar 1 measures for real estate exposures to address systemic risks? If you see a need, how should their coordination be strengthened?
Yes, we fully agree that there is a need. Currently, different authorities use overlapping requirements regarding the real estate portfolios (and some authorities for the non-performing portfolios).
There should be consistent principles for real estate exposures across all member states which allow the reflection of intrinsic national regulations (e.g. German covered bonds/Beleihungswert regulation). Furthermore, there should be more transparency (central publication) concerning in which countries strict(er) measures apply.
In addition, we believe that, for competitive reasons, a more harmonised pool of methods and transparency of the macro-prudential instruments used at national level would be called for.
Question 4: Do activity-based instruments in the current framework allow to effectively tackle risks stemming from specific risk exposures?
Yes, ESBG agrees with this statement as competent authorities can always use Pillar 2 add-ons to address any risks at micro- (idiosyncratic)-level as well as systemic risks. The use of Pillar 2 has not been subject to harmonisation, which leads to great discretion for competent authorities.
The more interesting question is how a certain tool is used, rather than if a proper tool is available. We think the existing legislation gives a consistent framework, and the issues for the banking sector actually stem from the way in which the different authorities implement it. The macro-prudential toolkit, in ESBG’s understanding, is flexible enough at the moment.
At the same time we would like to emphasise that it is impossible to make a clear distinction between institution-based and the activity-based tools. This is reflected in the classification of the tools in the consultation paper: the tools are typically mentioned in both categories. We would propose not using this grouping of the macro-prudential tools, as the distinction is rather relevant for the use/implementation of the tool, which can be both activity-based and institution-based in case of the same tool (it is not the tools that can be grouped, rather the implementation of the tools).
It is important to clarify the scope for each instrument in order to avoid a discretionary use. The purpose of the tools in the framework needs to be clearer in order to achieve a single rule book, which we do not have today. It is, in ESBG’s view, not always clear enough when the different tools could be used. Pillar 2 should preferably not be used for addressing systemic risk, but, if this was the case, such Pillar 2 requirements should be applied in the form of Pillar 2 guidance, or Pillar 2b. If they come last in the order, and are not accompanied by a formal decision procedure, they serve as a purposeful cushion, in a similar way as a bank’s own management buffer.
In respect of macro-prudential supervision, there is, in our view, no need to impose activity-based instruments, e.g. a countercyclical capital buffer specifically for residential real estate exposures, on banks. Mixing individual activity-based risks incurred by banks that are strongly active in a few specialised lines of business with an integrated macro-prudential approach to the financial market as a whole is not a convincing approach. Moreover, for such risks (e.g. heavier real estate business) enough macro-prudential tools are already available under, for example, the Standardised Approach for credit risk or the Mortgage Credit Directive.
Question 5: Do you consider a CCB for sectoral imbalances (e.g. in the real estate sector) a useful complementary instrument? If yes, how would you see the interaction of this sectoral CCB with the CCB already in place?
No. In our understanding, a sectoral imbalance indicates structural problems in the economy and can be dealt with by the SRB.
The concept underlying the CCB is to some extent questionable as it is, in principle, based on the loan-to-GDP ratio. There is no proof that it could be an effective instrument for dampening the credit cycle. The CCB does not necessarily resolve the pro-cyclicality problem, but could in fact exacerbate it. In bad economic times with low GDP growth the CCB signals the establishment of the additional capital buffer, and in good economic times with high GDP growth it does not. If a sector-specific CCB is now to be introduced in addition to the general CCB, this problem would be aggravated. What is more, how both instruments (general CCB and sector-specific CCB) unfold their effect and, in particular, how they interact with each other are not fully clear. This notwithstanding, a sectoral CCB could be too complex to implement for institutions, and would therefore not be advantageous in terms of transparency.
Particularly in the real estate sector, further regulatory action is not necessary in ESBG’s opinion. Hence, we are not in favour of the idea of a sectoral CCB.
Question 6: Do you see a need for adjusting measures targeting risks associated with banks’ real estate exposures? If so, please explain your answer.
No. In ESBG’s opinion, the regulatory framework is clear (Pillar 2 add-ons to cover institution-specific risks, SRB and Art 458 CRR to cover systemic risks) and gives sufficient, proper tools to the regulators. Instead, we suggest focusing on the consistent – amongst the tools and amongst the authorities - implementation of the existing tools.
Question 7: Do you see a need for disentangling different responsibilities between competent and designated authorities?
With uniform capital requirements in mind, we wish to draw the Commission’s attention to the following general problem that needs to be resolved for national real estate markets. Under Article 124(2) sentence 1 CRR and Article 164(5) sentence 1 CRR, the “competent authorities” periodically assess whether the risk weights for real estate exposures in the Standardised Approach for credit risk and the minimum LGD values for the internal-ratings-based approaches are appropriate.
For SSM institutions in the eurozone, the ECB is the “competent authority”, whilst in the case of LSIs the respective national authorities are responsible for deciding on any increases. The methodical approach to these two instruments allows the competent authorities a large degree of discretion.
In ESBG’s view, it is therefore likely that, for example, the ECB and an NCA will end up at different assessments for certain real estate markets (or segments thereof). Even if both believe that an increase in capital requirements is needed, the amount of the increase they decide on does not necessarily have to be identical.
The result would be different capital requirements for identical exposures and thus for identical risks within national real estate markets in the eurozone. Bearing in mind the “same risk, same rules” principle, this should be avoided. Also for competitive reasons.
Question 8: Do you see merit in better distinguishing the activity-based from the institution-based instruments under Article 458 CRR, also in view of applicable activation procedure(s)?
No, a better distinction is not necessary, in our opinion. As explained above, the distinction should refer to the usage of the tools, not to the tools themselves. If a tool would be activated for different purposes by different authorities in different procedures, the inconsistencies would be a larger problem.
As mentioned above, in our understanding the distinction between the activity-based and institution-based instruments is not as clear-cut. Hence, ESBG cannot see a convincing reason to further distinguish with regard to the activation procedures.
Question 9: Do you see the need to better frame either the focus (targeted risks) or the scope of the SRB (i.e. applicability to the entire stock only or also to subsets of exposures)? If so, please explain your answer.
Yes. The SRB is used by different authorities in different manners. In many cases, it serves to reach a predefined – by the regulator – level of capital requirements (“ring fencing”). Therefore, we would appreciate the clarification that the SRB shall not be implemented to cover institution-specific risks, nor shall it be used as a ring-fencing tool.
Question 10: Should the SRB be explicitly defined as either an activity based or an institution specific tool? Please explain your answer.
If specified, the SRB should be an activity-based instrument. However, ESBG cannot see a clear reason for this classification of the tools (please also see our responses above).
Question 11: How do you assess the interactions of institution-specific instruments in the current framework?
Instruments are overlapping, which might result in the same identified risk being covered more than once. Clear and well-defined provisions for the instruments, such as Pillar 2 and SRB, would reduce the risk for such overlapping risk coverage.
An overall assessment, done by the competent authority, of the combined effect resulting from all buffer requirements is necessary, even if the competence for the different buffers lies with different authorities.
Question 12: How do you assess the main weaknesses of institution-specific instruments in the current framework?
They are used to cover – in many cases parallel to the activity-based tools – systemic issues.
Institution-specific requirements like O-SII buffers should be considered within the SREP decision and should not weaken the goals of the banking union. Therefore, the EU or the SSM should be treated as one jurisdiction for the purpose of assessing “cross-border” activity in the G-SIB and O-SII assessment process.
Question 13: Do you consider that the capital buffers for systemically important institutions are appropriately calibrated in the current framework?
Yes, we think so. ESBG believes that the current framework is properly – efficiently and consistently – designed. Its implementation could be improved though.
Question 14: Do you assess the caps of the G-SII and the O-SII buffers as appropriate?
With regard to O-SII buffers, we assess them to be appropriate and appreciate that the cap on the sub-consolidated level is linked to the parent bank’s requirement. We also find it logical that the G-SII buffer is higher than the O-SII buffer cap as this reflects the relevance of the institution.
Question 15: Do you think that the 2 percent cap for the O-SII buffer should be revised? If so, please explain your answer.
No, we do not see a reason for this. The current structure is logically build, only the implementation should be monitored and corrected if justified.
Question 17: Do you see a need for developing additional harmonized macro-prudential instruments? If yes, what type of new instrument would you deem necessary and why?
No, as indicated above, ESBG sees the design of the current structure as logical. The implementation could perhaps be improved, and the macro-prudential instruments could be used in a more harmonised and consistent way, in line with their current purposes.
Question 18: How do you assess the possibility for the ESRB to develop technical guidance on the use of non-harmonised instruments, for example via issuing recommendations? Would you see a specific type of instrument for which such an approach could be warranted and suitable?
The ESRB is very much influenced by the ECB and lacks parliamentary accountability.
For general policy reasons, ESBG has reservations about the introduction of a further legislative instrument in the form of ESRB “recommendations”. This could once again create the danger of legislators’ original objectives being diluted or actually tightened through subsequent additional specification at another level. Such instances have already been experienced due to the Level 2 and Level 3 legal acts by European Supervisory Authorities. Regulation in the financial services sector could, as a result, become even more unmanageable and bureaucratic. Particularly small and medium-sized banks feel they are being increasingly overburdened.
There is no (democratic) need to delegate additional regulatory powers and technical guidance to the ESRB.
Question 19: Do you consider the current hierarchy of instruments ('pecking order') as appropriate?
Currently, the driver for which instrument the responsible authority uses seems to be the activation procedure and how framed the instruments are. To use the Swedish example, an SRB of 3% has been complemented with a 2% add-on in Pillar 2. So the bank in practice needs to hold an SRB of 5% without the responsible authority having to go through the activation procedure applicable for an SRB above 3%. Currently, the use of Pillar 2 measures are not framed, and the use of Pillar 2 compared to other macro-prudential tools needs to be clarified. In order for the hierarchy of instruments and activation procedures to function, all the tools available need to be clearly framed.
Pillar 2 should preferably not be used for addressing systemic risk, but if it is, such Pillar 2 requirements should be applied in the form of Pillar 2 guidance.
Concerning the implementation of the Pillar 2 guidance, the picture, however, has become more complicated. The ECB requests ESBG members “to treat a failure to meet the Pillar 2 capital guidance as an early warning signal in connection with the recovery plan […]”.
ESBG disagrees with this interpretation. As the ECB states, the Pillar 2 guidance is to be held additionally to the Pillar 1 requirements, Pillar 2 add-ons and combined buffers. These buffers have been put in place to ensure banks have sufficient funds available in stress situations so as not to breach their Pillar 1 and Pillar 2 requirements, i.e. they should be usable before a bank enters a potential recovery situation. We, therefore, believe that setting early warning signals for recovery plans before buffers are breached is not justified by risk management considerations, nor in line with the regulatory objectives of the buffers.
In contrast, in the bank’s Risk Assessment System (RAS), the Pillar 2 guidance is considered after the Pillar 2 requirement, but before the combined buffer, thus treating the combined buffer as an ‘amber zone’ for risk management purposes. Recovery ‘early warning signals’ are calibrated at the ‘red’ RAS limit, as the bank considers that a recovery situation could only occur (but does not necessarily occur) when it moves outside of its risk appetite.
In EBSG’s point of view, the role and limits of Pillar 2 guidance should be properly defined and should not cause further inconsistencies in the requirements.
Question 20: Can overlaps in the tools’ scope facilitate the circumvention of control elements embedded in the activation mechanism? If you answer yes, please explain how.
Yes it can, as Pillar 2 is not perfectly framed and can be used for the same risks as the other macro-prudential tools. Pillar 2 can facilitate the circumvention of control elements embedded in the activation mechanism (please also see our answer to question 19).
Question 21: What adjustments, if any, would you suggest for the notification and activation requirements for the SRB?
The SRB – and all other buffer requirements – shall be communicated to the banks in due time, giving the banks the chance to prepare for the change. Therefore, we suggest a publication requirement of 12 months in advance. A bad example would be Decision Nr. IV/1/4947/29.12.2015 of the National Bank of Romania about the implementation of an O-SII buffer requirement in Romania as of 1.1.2016 (only 2 working days later).
ESBG would also appreciate an impact study be made publicly available and which properly justifies the decision in case of any or all changes in the macro-prudential toolkit. We would appreciate if the impact study would cover the interactions between the already implemented and the newly planned tools.
Notification/activation should only be performed if prior to this a binding consensus is found for its harmonised implementation.
To mitigate long-term, non-cyclical or macro-prudential risk that may pose a threat to the financial system, NCAs can order the establishment of a capital buffer for systemic risk. The EU institutions have to be notified or asked for their approval in advance. Such an order is not to be addressed to a single bank but to all banks or certain types or groups of banks. In particular, the capital buffer can only apply to risk exposures located in a Member State, or it can be ordered for institutions at group or solo level.
ESBG believes that the provisions that allow NCAs to order systemic risk buffers are too unspecific. This buffer is currently designed as a “catch-all variable” and is applied by supervisors in a discretionary and relatively non-transparent manner. We see this buffer’s interaction with other capital buffers as redundant and questionable (in its current form).
Question 22: What adjustments, if any, would you suggest for the notification and activation requirements for the measures under Article 458 CRR?
Similar to our response to question 21, we would appreciate an early and appropriate publication date.
A competent or designated authority should judge the measures in the overall context of capital requirements imposed (on national and supra-national level).
Question 23: What adjustments, if any, would you suggest for the notification and activation requirements for the CCB?
The CCBs should be consolidated and published on the ESRB homepage both for Member States and for third countries. Thus, a central publication by the authorities would be highly appreciated.
Please see also our reply to question 5.
Question 24: Do you see the risk that especially the O-SII buffer and the SRB could be used for ring-fencing purposes? If yes, what do you suggest to address this risk?
Yes, ESBG members have already had this experience in fact (please see question 9).
Question 25: How do you assess the shared responsibilities of the ECB/SSM and national authorities for macro-prudential policy within the Banking Union? In particular, do you think that the current asymmetry of powers conferred upon the ECB/SSM is appropriate?
There is a heterogeneous application of setting the O-SIIs buffer and SRB for national institutions across different SSM countries. The effect of the discretionary application of buffers is enforced by requiring banks to fulfil these buffers on top of the SREP requirement and, therefore, consider them for the Maximum Distributable Amount (MDA). Furthermore, the O-SIIs buffer and SRB have to be considered twice (loss absorption amount & recapitalisation) when calculating MREL. To sum up, the shared responsibility regarding the O-SIIs buffer and SRB, while counting towards MDA and MREL, creates an unlevel playing field across different SSM countries.
Question 26: How do you assess the coordination need between the different authorities involved? Do you see areas in which this coordination could be improved?
As stated above, the coordination could be better, and at the end of the day, a comprehensive and more consistent macro-prudential requirement should be implemented.
Question 27: Do you see need for amending the time periods of the notification process between national authorities and the ECB/SSM? What time limitations would you suggest?
What is really needed is a proper and timely notification towards the banking industry.
Question 28: Do you see need to broaden the scope for mandatory reciprocity in the CRR/CRDIV? If yes, for which instrument(s) do you see such a need?
No, ESBG does not see a need to do so.
> See the .pdf version
About ESBG – the European Savings and Retail Banking Group
ESBG brings together nearly 1000 savings and retail banks in 20 European countries that believe in a common identity for European policies. ESBG members represent one of the largest European retail banking networks, comprising one-third of the retail banking market in Europe, with 190 million customers, more than 60,000 outlets, total assets of €7.1 trillion, non-bank deposits of €3.5 trillion, and non-bank loans of €3.7 trillion. ESBG members come together to agree on and promote common positions on relevant regulatory or supervisory matters.
European Savings and Retail Banking Group – aisbl
Rue Marie-Thérèse, 11 ￭ B-1000 Brussels ￭ Tel: +32 2 211 11 11 ￭ Fax: +32 2 211 11 99
Info@wsbi-esbg.org ￭ www.wsbi-esbg.org
Published by ESBG. October 2016.