Response to consultation on draft RTS on IRRBB supervisory outlier tests

Go back

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.

With regards to the recalibrated lower bound for post- shock IR levels in the EVE SOT and NII SOT, we believe the proposed floor is too low and the implications of the change for both the EVE SOT and NII SOT are underestimated:
o The suggested floor is considered too low given what banks have observed historically, especially on the short term. A further lowering of the floor on the short end does not seem to be substantiated.
o Further, we understood that recalibrating the floor is mainly motivated by the observance of baseline bond yield points below the current floor, not by the relevant risk-free rates, who upon some minor exceptions, were respecting the floor in the past and current low-rate period. The EBA proposal could lead to a severe change in impact for EVE SOT and particularly in NII SOT, as the down shock more or less doubles for short tenors and increases with up to 90bps.
o Such severe change undermines the determination and the analysis of threshold levels for SOT NII, which were done based on QIS results that were still applying the old floor. So if this new floor is accepted, we would expect that the proposed levels for SOT NII thresholds also consider the relevant shock change at that time (i.e. dec 2020). Furthermore, the more extreme the scenarios imposed, the more behavioural models would also have to be adjusted so that results remain sufficiently realistic.
o In addition, it is generally agreed that below the -100 bps floor monetary policy becomes ineffective. Therefore, it is not realistic that this -100 bps would be breached on a longer term basis, and consequently, not needed to lower it further.
The draft RTS envisages to apply two supervisory parallel shock scenarios for SOT NII that apply instantaneously with magnitudes that are not consistent with facts. To better fit reality, it would make better sense to assume that the shocks are applied gradually over the considered 12 months horizon. Conscious of the operational complexities that banks may be confronted with such a gradual scenario, we recommend a more pragmatic solution by adopting a lower magnitude for the elected instantaneous shock.

We welcome the EBA’s proposal regarding the use of a one-year time horizon. The shorter the time horizon, the more accurate the SOT.

With regards to the constant balance sheet with current commercial margins for new business for the NII SOT, we would like to point out that NII is not only a risk measure, and one could question the choices that are made for the sake of comparability: In relation to the Pillar 3 disclosure requirements, it would be confusing to disclose a technical, methodological NII sensitivity calculation that doesn’t relate and potentially conflicts with internal estimations by banks. In general, information shared with the public should remain the responsibility of banks. Banks should have the flexibility to decide on the approach which is most meaningful when disclosing Pillar 3 IRRBB information.

Unless limited to fair value accounting through P&L, the market value changes in the calculation of the NII SOT would not be acceptable as it deviates from actual NII, would overlap with the EVE perspective and would be at odds with hedging. We propose to exclude from the market value changes in the time value of optional hedges, for example of a swaption, as it cannot always be taken into account in hedge accounting. These hedges would then cause extra volatility of the NII SOT, while from an economic point of view they are indeed closed to cover the interest rate risk, in a buy and hold balance sheet strategy. An SOT should not restrict the sector from using certain hedging instruments.

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?

With regards to the calibration of the test statistic for a large decline, we refer to Question 1, more specifically to the recalibrated lower bound for post- shock IR levels where it is indicated that the proposed change of floor would heavily impact the outcome of the peer analysis based on which the current proposed levels were derived.
Even though metric 2 would be a more logical approach, we note the following issues:
o Currency aggregation: with regards to the inclusion of administrative expenses in metric 2 as well as the currency aggregation, we would like to point out that Finrep items of F02.00 are only reported at the level of reporting currency of an institution. A currency breakdown is not a reporting obligation as such, and therefore its availability within institutions cannot be assumed. As a result, the calculation of metric 2 can only be done at the level of reporting currency, without the (very restrictive) currency aggregation.
o In addition, the outcome of metric 2 is dependent upon the difference of respectively the historical NII with the baseline NII/adverse NII. Given that the FinRep historical NII envisages all currencies, while the baseline NII is limited to material currencies (based upon contribution to the balance sheet, not to NII: theoretically 4% of the balance sheet can account for more than 10-20% of the NI), this classifies metric 2 as being a non-robust metric.
As a result, we would consider metric 1 for the definition of the large decline to be a more stable approach.

However, the sector believes further analyses should be developed over time before electing a specific ratio and its accompanying large decline thresholds. It is in any case crucial to record the threshold well in advance and to clearly communicate how it has been determined. Whichever option is finally determined, we notice that a large decline in itself is no reason to impose supervisory measures.

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?

Febelfin would like to share 2 additional observations:
o We are silently moving to the most restrictive out of the former 2 early warning indicators: the draft RTS only refers to delta EVE/Tier 1 above or below -15% to define outliers as there is no mentioning anymore of the EVE/Total Capital above or below -20%.
o In addition, we note an undesirable evolution with regards to consequences of a potential breach. Direct actions are immediately expected (unless when the NCA considers that such actions would not be necessary) whereas before, the consequence was limited to informing the NCA after which potential actions could still be taken when considered reasonable. We would like to stress that the ALM of the bank should remain its own responsibility.

Upon final publication in the Official Journal of the European Union, the RTS should, in any case, enter into force no sooner than the end of the next calendar semester after publication date so as to enable banks to adapt. Ideally, the date of application should be aligned with the Guideline and the RTS on Standardized Methodologies.

Name of the organization

Belgian Financial Sector Federation (Febelfin)