Response to consultation on draft RTS on IRRBB supervisory outlier tests
Go back
Should nonetheless the EBA decide to proceed with the proposed recalibration of the lower bound, the effect of such change should be factored in the calibration of the NII SOT threshold. The proposed change would indeed heavily affect the resulting risk measure.
As an example, it can be simply noted that, for immediate maturity, starting from a baseline interest rate of around -50 bp, the shock applying the current lower bound (-100 bp) is 50bp, while applying the recalibrated floor (-150bp) the shock would be 100 bp (i.e., the shock is doubled).
A change in the lower bound would therefore imply an update of the quantitative study for the calibration of the NII SOT threshold (see also the response to Question 2).
As regards the market value changes, in ABI’s opinion they should be excluded from the calculation of the NII SOT, whose scope should be limited to the commonly used definition of NII, i.e. interest income and expenses.
Inclusion of other components, and namely changes in fair value, appears as a deviation from the CRD mandate, which explicitly refers to NII, as well as from the BCBS Standards.
Besides, consideration of market value changes in this context would not be consistent with the way the inherent risk is actually managed in banks.
Not less important, it would determine overlapping between NII measures and Economic Value (EV) measures, with a double counting of the fair value changes in the first year of the time horizon.
Last but not least, the significant differences among the fair value models applied by each bank would negatively affect the comparability of the results of the NII SOT (also given that the impact of this component would be material).
For all the above reasons, ABI would recommend excluding changes in fair value from the NII SOT metric (see also the response to Question 2 below).
As regards the other assumptions, ABI agrees that the most appropriate time horizon for the NII SOT is one year. It is considered appropriate in principle, and in addition it should be considered that the longer the time horizon, the less realistic would be the constant balance sheet assumption.
Metric 2, instead, appears as not representative of IRRBB and detached from risk management practices, as it involves administrative expenses, which is an item not related to IRRBB and outside the remit and control of ALM/risk management units.
Moreover, Metric 2 would be less transparent, and its results would be complex to explain, both internally and outside the bank (to market analysts). Also, ongoing monitoring would be more complex, as updated accounting figures might not always be easily available.
In addition, administrative expenses are heavily affected in case of extraordinary operations (e.g. M&A) and events. This would imply the need for a framework to manage extraordinary circumstances (while ensuring comparability of results).
Given the above, ABI has a clear preference for Metric 1.
Anyway, ABI would highlight that the appropriateness of the metric should be considered in conjunction with the calibration of the threshold.
In this regard, ABI is concerned that the proposed calibration is overly strict. This seems to be the result of the assumptions of the quantitative study, which only considered the current very low interest rate environment.
Since the SOT is expected to be applied in the medium to long term, a proper assessment should take into account that the threshold will likely be applied in “normal” times with a positive interest rates environment, when the impact of the interest rate shocks on NII will be much less constrained by legal, contractual and regulatory floors.
It has to be noted that the QIS exercise included “unconstrained” figures, but these are not considered fit for purpose, as they were to be determined disapplying the regulatory lower bound only. The assessment should instead consider the impact of the full shock, as well as possible difference in behavioural assumptions under different circumstances.
A calibration of the threshold, based on available figures only, would hence represent a fatal flaw.
In addition, banks have been requested to provide figures for the quantitative study based on the lower bound currently applied to interest rates after shocks. As said in the response to Question 1, should the envisaged recalibration of the lower bound be confirmed in the final RTS - which ABI does not recommend – this would itself imply the need for reassessing the calibration of the threshold. Evidence from analysis conducted by some Italian banks shows that the mere change of the lower bound would determine banks crossing the threshold.
ABI would therefore urge the EBA to revive the quantitative study, in order to properly calibrating the NII SOT threshold. Should this not be possible, the threshold should at least be conveniently adjusted to account for the abovementioned aspects.
As a matter of fact, the implementation of the new NII SOT will imply adaptation of banks’ strategies and policies, also in light of the fact that it will be applied in conjunction with the existing SOT in the economic value perspective (EVE SOT), and the combined effects of the two will have to be managed. A too strict threshold would therefore severely constrain banks’ room for manoeuvre and ultimately not only unduly affect banks’ profitability, but also limit the supply of products and services to their clients.
In this regard, ABI would highlight a key point. It is well established in the Directive - but it might be worth stressing further in the RTS - that crossing the threshold shall not automatically entail supervisory measures. The latter should be applied only following to a case-by-case assessment of a bank’s interest rate risk management. Anyway, such flexibility shall not lead to believe that a too strict calibration of the threshold would have no or little consequences.
Indeed, banks’ strategies and policies will be shaped considering that threshold as binding. Therefore, although crossing the threshold might not always determine direct consequences on a bank in terms of supervisory measures (ex-post), nonetheless all banks will be negatively affected (ex-ante) due to the constraint that an overly conservative threshold would pose to their operations.
That is why ABI remarks that an appropriate calibration of the threshold is key and recommends establishing in any case a two-years observation period after first definition of the threshold (envisaging a possible review of its calibration).
As to the other points addressed under this question, ABI would consider beneficial further clarification of the rules for the currency aggregation, as the proposed approach appears not straightforward, especially as regards the combination of the results of different pairs of currencies and the treatment of ERMII currencies compared to EUR.
It might be worth reminding that the legislator, in CRD5, had conveniently stipulated a one-year span between availability of the regulation (June 2020) and its application (June 2021).
Indeed, as highlighted above, the new regulations will not only require implementation but will also imply changes in banks’ strategies and internal policies, which in turn require strong governance (approval processes, audit and Board/top management involvement). Hence, the implementation of the new framework has to be integrated in a well-defined and rather rigid process. This requires time starting from the moment when banks have legal certainty on what they have to do, and when, since only to a limited extent the required actions can be planned in advance, in the absence of legal certainty.
This means that a reasonable timing for implementation is needed, and that such timing shall not be computed starting from the moment when the EBA delivers the final draft RTS, but from the publication on the OJEU.
Respondents are also kindly requested to express whether they find an inclusion of market value changes in the calculation of the NII SOT clear enough.
In ABI’s opinion the proposed recalibration of the lower bound, to be applied to post-shock interest rate levels, is not needed. The current lower bound is considered appropriate, having it proven fit (with minor exceptions) in the current interest rate scenario, and given that evolution towards a lower level is deemed highly unlikely. The recalibrated lower bound seems therefore rather unrealistic.Should nonetheless the EBA decide to proceed with the proposed recalibration of the lower bound, the effect of such change should be factored in the calibration of the NII SOT threshold. The proposed change would indeed heavily affect the resulting risk measure.
As an example, it can be simply noted that, for immediate maturity, starting from a baseline interest rate of around -50 bp, the shock applying the current lower bound (-100 bp) is 50bp, while applying the recalibrated floor (-150bp) the shock would be 100 bp (i.e., the shock is doubled).
A change in the lower bound would therefore imply an update of the quantitative study for the calibration of the NII SOT threshold (see also the response to Question 2).
As regards the market value changes, in ABI’s opinion they should be excluded from the calculation of the NII SOT, whose scope should be limited to the commonly used definition of NII, i.e. interest income and expenses.
Inclusion of other components, and namely changes in fair value, appears as a deviation from the CRD mandate, which explicitly refers to NII, as well as from the BCBS Standards.
Besides, consideration of market value changes in this context would not be consistent with the way the inherent risk is actually managed in banks.
Not less important, it would determine overlapping between NII measures and Economic Value (EV) measures, with a double counting of the fair value changes in the first year of the time horizon.
Last but not least, the significant differences among the fair value models applied by each bank would negatively affect the comparability of the results of the NII SOT (also given that the impact of this component would be material).
For all the above reasons, ABI would recommend excluding changes in fair value from the NII SOT metric (see also the response to Question 2 below).
As regards the other assumptions, ABI agrees that the most appropriate time horizon for the NII SOT is one year. It is considered appropriate in principle, and in addition it should be considered that the longer the time horizon, the less realistic would be the constant balance sheet assumption.
Question 2: Do respondents have any comment related to these two metrics for the specification and the calibration of the test statistic for the large decline in Article 6 for the purpose of NII SOT? Specifically, do respondents find the inclusion of administrative expenses in metric 2 clear enough? Do respondents have any comment on the example on currency aggregation for metric 1 and metric 2?
As regards the two possible metrics for the NII SOT proposed in the draft RTS, Metric 1 (Tier 1 as denominator) is simple and clear.Metric 2, instead, appears as not representative of IRRBB and detached from risk management practices, as it involves administrative expenses, which is an item not related to IRRBB and outside the remit and control of ALM/risk management units.
Moreover, Metric 2 would be less transparent, and its results would be complex to explain, both internally and outside the bank (to market analysts). Also, ongoing monitoring would be more complex, as updated accounting figures might not always be easily available.
In addition, administrative expenses are heavily affected in case of extraordinary operations (e.g. M&A) and events. This would imply the need for a framework to manage extraordinary circumstances (while ensuring comparability of results).
Given the above, ABI has a clear preference for Metric 1.
Anyway, ABI would highlight that the appropriateness of the metric should be considered in conjunction with the calibration of the threshold.
In this regard, ABI is concerned that the proposed calibration is overly strict. This seems to be the result of the assumptions of the quantitative study, which only considered the current very low interest rate environment.
Since the SOT is expected to be applied in the medium to long term, a proper assessment should take into account that the threshold will likely be applied in “normal” times with a positive interest rates environment, when the impact of the interest rate shocks on NII will be much less constrained by legal, contractual and regulatory floors.
It has to be noted that the QIS exercise included “unconstrained” figures, but these are not considered fit for purpose, as they were to be determined disapplying the regulatory lower bound only. The assessment should instead consider the impact of the full shock, as well as possible difference in behavioural assumptions under different circumstances.
A calibration of the threshold, based on available figures only, would hence represent a fatal flaw.
In addition, banks have been requested to provide figures for the quantitative study based on the lower bound currently applied to interest rates after shocks. As said in the response to Question 1, should the envisaged recalibration of the lower bound be confirmed in the final RTS - which ABI does not recommend – this would itself imply the need for reassessing the calibration of the threshold. Evidence from analysis conducted by some Italian banks shows that the mere change of the lower bound would determine banks crossing the threshold.
ABI would therefore urge the EBA to revive the quantitative study, in order to properly calibrating the NII SOT threshold. Should this not be possible, the threshold should at least be conveniently adjusted to account for the abovementioned aspects.
As a matter of fact, the implementation of the new NII SOT will imply adaptation of banks’ strategies and policies, also in light of the fact that it will be applied in conjunction with the existing SOT in the economic value perspective (EVE SOT), and the combined effects of the two will have to be managed. A too strict threshold would therefore severely constrain banks’ room for manoeuvre and ultimately not only unduly affect banks’ profitability, but also limit the supply of products and services to their clients.
In this regard, ABI would highlight a key point. It is well established in the Directive - but it might be worth stressing further in the RTS - that crossing the threshold shall not automatically entail supervisory measures. The latter should be applied only following to a case-by-case assessment of a bank’s interest rate risk management. Anyway, such flexibility shall not lead to believe that a too strict calibration of the threshold would have no or little consequences.
Indeed, banks’ strategies and policies will be shaped considering that threshold as binding. Therefore, although crossing the threshold might not always determine direct consequences on a bank in terms of supervisory measures (ex-post), nonetheless all banks will be negatively affected (ex-ante) due to the constraint that an overly conservative threshold would pose to their operations.
That is why ABI remarks that an appropriate calibration of the threshold is key and recommends establishing in any case a two-years observation period after first definition of the threshold (envisaging a possible review of its calibration).
As to the other points addressed under this question, ABI would consider beneficial further clarification of the rules for the currency aggregation, as the proposed approach appears not straightforward, especially as regards the combination of the results of different pairs of currencies and the treatment of ERMII currencies compared to EUR.
Question 3: Do respondents consider that all the necessary aspects have been covered in the draft regulatory standard? Do respondents find the provisions clear enough or would any additional clarification be needed on any aspect?
The proposed entry into force of the RTS only 20 days after publication on Official Journal raises concerns.It might be worth reminding that the legislator, in CRD5, had conveniently stipulated a one-year span between availability of the regulation (June 2020) and its application (June 2021).
Indeed, as highlighted above, the new regulations will not only require implementation but will also imply changes in banks’ strategies and internal policies, which in turn require strong governance (approval processes, audit and Board/top management involvement). Hence, the implementation of the new framework has to be integrated in a well-defined and rather rigid process. This requires time starting from the moment when banks have legal certainty on what they have to do, and when, since only to a limited extent the required actions can be planned in advance, in the absence of legal certainty.
This means that a reasonable timing for implementation is needed, and that such timing shall not be computed starting from the moment when the EBA delivers the final draft RTS, but from the publication on the OJEU.
Upload files
PP_ABI_IRRBB_def SOT.pdf
(161.14 KB)