Response to consultation on draft Guidelines on IRRBB and CSRBB
Go back
The French Banking Federation (FBF) welcomes the opportunity to express the views of the French banking industry on the public consultation on the draft Guidelines on the identification, evaluation, management and mitigation of risks arising from potential changes in Interest Rates and of the assessment and monitoring of Credit Spread risk, of institutions’ non-Trading Book activities. In this context, we herewith provide you with our general remarks and responses to the questions listed in the Consultation Paper (CP). We appreciate your consideration about our comments and remain at your disposal for further clarifications.
Article 84(6) of Capital Requirement Directive (CRD) tasks the European Banking Authority (EBA) with developing a Guideline to define:
• the evaluation by an institution's internal system;
• the identification, management and mitigation by institutions of the risks;
• the assessment and monitoring by institutions of the risks;
• determining which of the internal systems are not satisfactory.
The European Banking Authority (EBA) has published Guidelines on Interest Rate Risk in the Banking Book (IRRBB) in May 2015 and July 2018 and is to publish yet another version of its Guideline on the very same topic in the course of 2022. There is a need to a stable regulatory framework and we are hopeful that it will be stabilized.
In July 2018, EBA elected to modify its May 2015 Guideline to factor in Basel Committee on Banking Supervision (BCBS) Standard on IRRBB published in April 2016 ahead of the mandate to be received from CRR to factor in this very same BCBS Standard. Hence, the EBA 2018 Guideline should be only marginally modified.
However, the Consultation Paper (CP) introduces very significant deviations from July 2018 Guideline that are not substantiated by any change in BCBS or CRR mandate and actually can be considered as:
• the CP envisages to change the definition of commonly understood Net Interest Income (NII) to include changes in fair values of instruments even though they are not part of NII. This appears as a deviation from the CRR mandate that explicitly refers to NII. It also deviates from BCBS Standards that is quite explicit that it refers to NII excluding changes in fair values that don’t affect NII. This would also be at odds with actual risk management and would introduce overlapping between NII measures and Economic Value (EV) measures while they should be complementary. NII should be kept as defined by interest income and expenses.
• The CP envisages a new limitation to the duration of Non-Maturing Deposits (NMD’s) (i.e. a 5 year cap on NMD’s from retail and non-financial wholesale, and a 0 year cap on NMD’s from financial customers) following a said “prudent” approach. These caps are inconsistent with a prudent approach since interest rate risk is a symmetric risk. Without any economic justification, they would lead to inappropriate measurement and management if incorporated in a management framework. It is reminded that, such caps are neither in CRR/CRD nor in BCBS Standard (where such caps appear only in the non-mandatory standardized method). Furthermore, the mandate to the EBA on the definition of SOT explicitly excludes behavioral assumption (CRD.Art.98(5a): ‘common modelling and parametric assumptions, excluding behavioral assumptions), the introduction of caps on NMD’s in the general framework of IRRBB management appears as EBA working around the mandates that it received.
Caps on NMD’s should be deleted.
• the CP envisages dramatic changes to the definition and scope of Credit Spread Risk in the Banking Book (CSRBB) while the July 2018 EBA Guideline already implemented the BCBS Standard that has not changed since then. The envisaged changes are not only not substantiated but they would also introduce significant confusions and complexities. CSRBB should be unchanged compared to July 2018 EBA Guideline, notably with a scope defined by assets that are actively traded on a deep and liquid markets to have identifiable and measurable market perception and changes thereof.
CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation are limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
• the CP provides principles for the assessment by competent authorities of the management framework for IRRBB and CSRBB. Broadly, those principles make sense to cover the different dimensions of the management framework. However, to the extent that the assessment could lead competent authorities to require banks to apply the Standardized Methodology for the evaluation of IRRBB, it is essential to better frame the conditions under which banks such a requirement could apply.
The requirement to apply the Standardized Methodology for the evaluation should relate to the evaluation component of the Internal Management System (IMS) that should be evidenced as non-satisfactory. In other words, a framework that would be assessed as non-satisfactory for issues that are not related to its evaluation component of its IMS should not lead to require using the Standardized Methodology.
As any Standardized Methodology would not be risk sensitive, as made clear by BCBS, it should be clarified that requiring its application should follow a careful analysis to evidence whether the Standardized Methodology would be more risk sensitive than the IMS evaluation it would substitute. Bar such an evidence, Standardized Methodology should not apply. Moreover, the GL should require competent authorities to grant institutions the possibility to remedy potential shortcomings in their IMS within a reasonable period of time before the Standardized Methodology is imposed. Last but not least, if the Standardized Methodology was imposed after all, this should only be temporary and a return to IMS should be allowed as soon as possible.
• We would like to stress that the implementation of the Guideline will require a significant time to be carried out. Consequently, a sufficient period (at least 1 year) should be assigned before application. Ideally, the final application date of the Guideline should be aligned with the application of the technical standards, as the EBA suggested already in article 8.
There are many other necessary changes to the draft Guideline to make it appropriate and consistent, notably on the suggested definitions (many of which changes since July 2018 EBA Guideline!) and the notion of ‘prudence’. Interest rate risk is a symmetrical risk: it is as risky to adopt too short durations than too long durations when considering banking book items. The CP is biased in considering prudence as leaning towards shorter duration, which does not make economic sense.
The changes are presented in detail below.
We are committed to elaborate a sound and robust management framework for IRRBB and CSRBB and remain available to EBA to support its efforts.
Response to Question 1 :
• The CP envisages a new limitation to the duration of Non-Maturing Deposits (NMD’s) (i.e. a 5 year cap on NMD’s from retail and non-financial wholesale, and a 0 year cap on NMD’s from financial customers) following a said “prudent” approach. These caps are inconsistent with a prudent approach since interest rate risk is a symmetric risk. Without any economic justification, they would lead to inappropriate measurement and management if incorporated in a management framework. It is reminded that, such caps are neither in CRR / CRD nor in BCBS Standard (where such caps appear only in the non-mandatory standardized method). Furthermore, the mandate to the EBA on the definition of SOT explicitly excludes behavioral assumption (CRD.Art.98(5a): ‘common modelling and parametric assumptions, excluding behavioral assumptions), the introduction of caps on NMD’s in the general framework of IRRBB management appears as EBA working around the mandates that it received.
Caps on NMD’s should be deleted.
There should be no limitation to the development of behavioral models that are instrumental in the risk-sensitive management of IRRBB. In particular, there should be no limitation to the maturity of deposits.
All of the customer-related Banking Book (BB) items require behavioral models to convey the actual interest rate risk exposure. Such models are essential to factor in the idiosyncrasies of each product, product mix, customer, competition, business model (including customer relationship, product mix), regulation, law and tax environment. Those idiosyncrasies need to be considered in the measurement, monitoring and management of IRRBB.
This has been recognized by BCBS in its Standard published as recently as 2016 (confirming the previous 2004 BCBS Standard) and confirmed in the CRD V Article 98.5a which clearly excludes from the EBA mandate the possibility to specify assumption in relation with behavioral assumptions for SOT calculus. As a consequence, the proposed five-year cap should be considered as both out of the mandate of EBA and detrimental to the quality of the IRRBB management framework of European banking institutions.
The statement that the EBA proposal is ‘less strict than the Basel standardized approach’ (p41 of the explanatory box of CP 2021/37) is inconsistent and misleading as it wrongly compares the management framework based on internal models that the BCBS fully recognizes without limitation to NMD’s. Actually, the BCBS Standard only suggests a standardized approach limited to be a fallback approach in case of non-satisfactory internal approach and in this fallback approach, NMD’s are clearly described as non-amenable to standardization. The introduction of any binding hard-coded limitation in the internal models would actually be a deviation from the BCBS Standard.
Any binding hard-coded limitation to the behavioral models would lead to failing to represent the actual interest rate risk exposure, would lose the fundamental risk-sensitivity of the IRRBB measurements and result in mitigating transactions that could actually magnify the interest rate risk exposure rather than mitigate it. It hardly can be a regulatory or supervisory objective to impose a fundamentally flawed measurement, monitoring and management of interest rate risk that it is supposed to regulate / supervise. In such binding hard-coded limitation would be a fatal flaw in the regulatory / supervisory framework.
For instance, the natural offsets with other interest rate risk exposures in the BB (e.g. fixed interest rate loans, notably mortgages) would not be recognized, leading to apparent and wrong interest rate risk exposure. This would lead to entering into mitigation derivative transactions to ‘hedge’ those flawed interest rate risk exposures that would actually increase the interest rate risk exposure. This could also lead banks to limit the extension of long-term fixed rate loans that their customers demand, exposing customers to changes in interest rate risk (as they will be exposed to increase in interest rates when repricing / rolling over their loans). None of this makes sense.
As an illustration, French customers demand overwhelmingly fixed interest rate mortgages that are naturally offset by a significant portion of French customers non-interest-bearing demand deposits whose stability has been evidenced over very long-term period and through very different interest rate environment.
This also applies to specific products such as Plan d’Epargne Logement (PEL) which is a long dated fixed rate savings accounts and to regulated savings accounts (e.g. Livret A) which exist since the 19th century.
Some extracts from BCBS Standard relating to NMD’s are worth reminding, while reminding that BCBS does not prescribe any cap on NMD’s neither in the management framework recommended by BCBS, nor in the SOT (Principle 10):
§3. The Committee noted the industry’s feedback on the feasibility of a Pillar 1 approach to IRRBB, in particular the complexities involved in formulating a standardised measure of IRRBB which would be both sufficiently accurate and risk-sensitive to allow it to act as a means of setting regulatory capital requirements. The Committee concludes that the heterogeneous nature of IRRBB would be more appropriately captured in Pillar 2.
Process for positions not amenable to standardisation […] §109. Positions not amenable to standardisation include (i) NMDs […]
The perception that it is prudent to be shorter than longer does not make any economic or supervisory sense. Interest rate risk is a symmetrical risk: being too short is as risky as being too long.
For these reasons, we urge EBA to not impose any such hard-coded limitation to behavioral model and to delete paragraph 111.
Indeed, the paragraph below should be deleted:
§111. The assumed behavioural repricing date for retail and non-financial wholesale deposits without any specific repricing dates (non-maturity deposits) should be constrained to a maximum weighted average repricing date of 5 years. The 5-year cap applies individually for each currency. Non-maturity deposits from financial customers should not be subject to behavioural modelling.
More specifically, internal systems should be considered as not satisfactory in the following cases at least:
a) An IMS should be considered non-satisfactory for the purposes of paragraph 3 of Article 84 CRD if competent authorities assess, on a case by case basis, that the implemented methods do not cover all the material components of the interest rate risk (gap risk, basis risk, option risk), and/or measures do not capture in a robust and economically justified manner all material dimensions of risks for significant assets, liabilities and off-balance sheet type instruments (e.g. NMD, loans, options) of the bank’s non-trading book.
Annex I describes a non-restrictive list of methods for IRRBB measurement with an indication of their limitations.
When measuring their exposure to IRRBB, institutions should not limit themselves to the methods listed in Annex I in order to ensure that material aspects of interest rate risk are captured adequately.
b) IMS should be considered as non-satisfactory if they are not calibrated, back tested and reviewed in all their relevant parameters on an appropriate frequency and supported by a due governance and documentation that considers the nature, scale and complexity of the IRRBB inherent in the business model and the institution's activities.
Institutions should duly comply with para. 71 to 79 of these Guidelines in particular with review and validation (at their appropriate frequency, including back testing), governance, risk policies as well as controls.
Broadly, those statements make sense. However, experience shows that supervisor may apply some of EBA Guideline at the letter without applying proportionality principle. As the consequences of being considered non-satisfactory are important, notably with the use of the Standardized Approach that will fail to accurately represent interest rate risk exposure and will lead to wrong interest rate risk management decision, an internal system should be assessed as non-satisfactory only when the non-satisfactory points are evidenced as being material for measurement and management.
The paragraph 119(b) appears too detailed as the modeling framework is already referred to §71-79,. It is also potentially misleading as it may be understood as requiring each single parameter to be subject to back-testing rather than models as a whole, and as some assumptions are necessarily discretionary (e.g. choice of the duration of equity).
We propose amendments to §119 like below:
More specifically, considering the nature, scale and complexity of the IRRBB inherent in the business model and the institution's activities and applying proportionality principle in the assessment, internal systems should be considered as not satisfactory in the following cases:
a) An IMS should be considered non-satisfactory for the purposes of paragraph 3 of Article 84 CRD if competent authorities assess, on a case by case basis, that the implemented methods do not cover all the material components of the interest rate risk (gap risk, basis risk, option risk), and/or measures do not capture in a robust and economically justified manner all material dimensions of risks for significant assets, liabilities and off-balance sheet type instruments (e.g. NMD, loans, options) of the bank’s non-trading book.
Annex I describes a non-restrictive list of methods for IRRBB measurement with an indication of their limitations.
When measuring their exposure to IRRBB, institutions should not limit themselves to the methods listed in Annex I in order to ensure that material aspects of interest rate risk are captured adequately.
b) IMS should be considered as non-satisfactory if they lead to material deviation from the application of para. 71 to 79 of these Guidelines in particular with review and validation, governance, risk policies as well as controls.
In addition, it should be made clear that the requirement to apply the Standardized Methodology for the evaluation should relate to the evaluation component of the Internal Management System (IMS) that should be evidenced as non-satisfactory. In other words, a framework that would be assessed as non-satisfactory for issues that are not related to its evaluation component of its IMS should not lead to require using the Standardized Methodology.
As any Standardized Methodology would not be risk sensitive, as made clear by BCBS, it should be clarified that requiring its application should follow a careful analysis to evidence whether the Standardized Methodology would be more risk sensitive than the IMS evaluation it would substitute. Bar such an evidence, Standardized Methodology should not apply. Moreover, the GL should require competent authorities to grant institutions the possibility to remedy potential shortcomings in their IMS within a reasonable period of time before the Standardized Methodology is imposed. Last but not least, if the Standardized Methodology was imposed after all, this should only be temporary and a return to IMS should be allowed as soon as possible.
CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation is limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
We consider that CSRBB guidelines are quite unclear. More specifically, the different spread components definitions are not so clear and the practical procedure to identify them neither.
See Response to Question #5 for more details.
CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation is limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
The proposed perimeter is much larger than the one currently considered by European and French banks. Furthermore, the methodology to be applied is not clear for us.
Cf. Response to Question #5 for more details.
• the CP envisages dramatic changes to the definition and scope of Credit Spread Risk in the Banking Book (CSRBB) while the July 2018 EBA Guideline already implemented the BCBS Standard that has not changed since then. The envisaged changes are not only not substantiated but they would also introduce significant confusions and complexities. CSRBB should be unchanged compared to July 2018 EBA Guideline, notably with a scope defined by assets that are actively traded on a deep and liquid markets to have identifiable and measurable market perception and changes thereof.
CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation are limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
As EBA elected to factor the BCBS 2016 Standard in its 2018 EBA Guideline, ahead of the finalization of the CRR, and as there was no other elements published by BCBS, we expect EBA Guideline to remain unchanged compared to 2018 EBA Guideline with the definition of CSRBB:
Definition (p14): The risk driven by changes in the market perception about the price of credit risk, liquidity premium and potentially other components of credit-risky instruments inducing fluctuations in the price of credit risk, liquidity premium and other potential components, which is not explained by IRRBB or by expected credit/(jump-to-) default risk.
Scope (p17): Institutions should monitor and assess their CSRBB-affected exposures, by reference to the asset side of the non-trading book, where CSRBB is relevant for the risk profile of the institution.
In the CSRBB perimeter defined in §120, it is understood and agreed upon that CSRBB relates to items that are subject to a market perception, i.e. with sufficiently breadth and depth market. Hence, BB items that are not subject to a market, e.g. loans and deposits, are not part of CSRBB. This should be explicitly mentioned.
In §121, it is mentioned that ‘the deterioration of an institution’s credit quality should not have any positive impact on the credit spread risk measure’. As it would make no sense to consider that the improvement of an institution’s credit quality should be considered as a risk for the institution, it leads to the conclusion that credit spread risk can only relate to the asset side… as EBA concluded in its EBA 2018 Guideline.
We recommend that the scope of application of CSRBB is confirmed as covering the asset side of the non-trading book, as EBA 2018 Guideline and to mention explicitly that Loans and Deposits to be excluded from this scope.
Actually, CSRBB should be consistent with the business model in which those items are held. To the extent that those items, subject to a deep market, are not intended to be sold before their maturity, those items should typically not be considered in the scope of CSRBB. This applies to market-liquid debt securities held in a Collect mode (and consequently accounted for without any impact of their changes in market values). This is consistent with §152 (Institutions should implement robust internal measurement systems (IMSs) that capture all components and sources of CSRBB which are relevant for the institution’s business model.).
As hedging derivatives in the BB are usually collateralized (directly or indirectly by their internal trading department), and subject to counterparty risk framework, they would not be subject to CSRBB except if they mitigate CSRBB.
We recommend deleting the explanatory box ‘Examples of CSRBB identification and measurement’ as it is too prescriptive and too complex:
• economic value perspective does not mean necessarily Economic Value of Equity (EVE)
• the breakdown by rating may not be the most appropriate (e.g. breakdown by issuer type may be more relevant)
• the breakdown by terms appears too prescriptive and complex
As a conclusion, the EBA Guideline should be modified:
Paragraph §124 should be modified from:
§124. Institutions should not exclude any instrument in the banking book from the perimeter of CSRBB ex ante, including assets, liabilities, derivatives and other off-balance sheet items such as loan commitments, irrespective of their accounting treatment. Any potential exclusion of instruments from the relevant perimeter should be done in the case of the absence of sensitivity to credit spread risk and should be appropriately documented and justified. In any case, institutions should not exclude assets accounted at fair value.
… to:
§124. Institutions should include the banking book assets that are actively traded on a deep and active markets (i.e. Level 1 assets) in the perimeter of CSRBB held in a collect and sell a business model. Any potential exclusion of instruments from the relevant perimeter should be done in the case of the absence of sensitivity to credit spread risk and should be appropriately documented and justified. In any case, institutions should not exclude assets accounted at fair value.
In terms of governance, the management body should be able to delegate its responsibility for CSRBB as for IRRBB:
Paragraph §130(a) should be aligned on §41(a):
§130(a) That their management body bears the ultimate responsibility for the oversight of the CSRBB management framework and the institution’s risk appetite framework to adequately cover the risks. The management body may, however, delegate the monitoring and management of IRRBB to senior management, expert individuals or an asset and liability management committee under the conditions further specified in paragraph 132.
Then, the management body should be systematically accompanied with ‘or its delegates’ (as in §131(e), §135, §136, §145, §146, §148, §149, §151):
§131 The management body or its delegates […]
§131(d) […] The level of and changes ni the institution’s CSRBB exposure should be provided regularly to the management body or its delegates.
§131(f) Understanding and assessing the functioning of its delegates in monitoring and controlling CSRBB, consistent with policies approved by the management body or its delegates, on the basis of regular reviews of timely and sufficiently detailed information.
§133 The management body or its delegates […]
The articulation and separation between CSRBB and IRRBB need to be clarified as it was in the EBA 2018 Guideline: it should be clear that CSRBB excludes IRRBB.
The exclusion of idiosyncratic components would make CSRBB far more complex as market perceptions (ability to retrieve prices from an organized liquid market) do not segregate the idiosyncratic component. Isolating it would be theoretical as methods would need to be developed with introducing models. The discretion that EBA CP envisages in §157 to include the idiosyncratic component subject to being more conservative, would make it even more complex (§157: As an exception, in the practical implementation of paragraph 120 and for proportionality reasons, institutions may include idiosyncratic credit spread components for the monitoring of CSRBB, as long as it is ensured that the measures will yield more conservative results.) and would not ease practical implementation.
Moreover, with the implementation of the Benchmark Regulation, the market indicator that represented the liquidity premium of the AA banks that was the Libor will disappear for most major currencies with the current exception of EUR. It will no longer be possible to differentiate between purely idiosyncratic and systemic premiums.
We recommend that CSRBB is defined without exclusion of sub-components such as idiosyncratic.
Finally, we propose to delete §131(g) as it would apply to any single type of risk and there is no reason to state this specifically for CSRBB.
An extract from BCBS Standard relating to CSRBB are worth reminding as they clearly highlight that it relates to items that are at fair value (mark-to-market), as illustrated by the graph (re. red zone) below:
Annex 1 - §5 - Changes to the market liquidity spreads and market credit spreads are combined within the definition of CSRBB. The diagram below gives a visual representation of how the various elements fit together.
• the CP envisages to change the definition of commonly understood Net Interest Income (NII) to include changes in fair values of instruments even though they are not part of NII. This appears as a deviation from the CRR mandate that explicitly refers to NII. It also deviates from BCBS Standards that is quite explicit that it refers to NII excluding changes in fair values that don’t affect NII. This would also be at odds with actual risk management and would introduce overlapping between NII measures and Economic Value (EV) measures while they should be complementary. NII should be kept as defined by interest income and expenses.
Net Interest Income (NII) means net interest income that is widely defined as the portion that impact the profit and loss statement.
Considering the changes in Other Comprehensive Income (OCI) and/or in capital as NII would be inconsistent with commonly accepted definition of NII, with CRD and with BCBS.
It would be so flawed as it would disincentivize to hedge with cash flow hedging instruments as their changes in fair value would be considered as a risk to NII while they are entered into precisely to make NII less sensitive.
We urge EBA to adhere to the common definition of NII, not to invent another definition of its own and to be consistent to the mandate provided by CRD that refers to Net Interest Income. EBA would deviate from its mandate by extending the definition of NII.
It is reminded that the economic value perspective has also to be considered and that the extension of NII measures would create overlap with EV measures.
Several paragraphs need to be fixed of this deviation like this:
15. For the purposes of these Guidelines, the net interest income upon which to calculate the impact of interest rate or credit spread movements should be determined by the interest income and expenses, and the market value changes of instruments — depending on accounting treatment in the profit and loss account. Institutions should take into account the increase or reduction in the amount of profit and losses over short- and medium-term horizons resulting from interest rate or credit spread movements.
§27(e) the impact on economic value and net interest income of mismatched positions in different currencies;
§31(d) The relative importance of interest rate sensitive instruments (including interest rate derivatives) in the non-trading book, with potential effects shown in the profit and loss account;
§44(c) In defining their risk appetites, institutions should take account of net interest income risks that may arise as a consequence of the accounting treatment of transactions in the non-trading book. The risk to net interest income may not be limited to interest income and expenses: the effects of changes in interest rates on the market value of instruments that, depending on accounting treatment, are reflected through the profit and loss account, should be taken into account. Institutions should particularly take into account the impact related to embedded optionalities in fair value instruments under ongoing interest rate shocks and stress scenarios. Institutions should also take into account the potential impact on the P&L accounts of hedging interest rate derivatives if their effectiveness was hampered by interest rate changes.
Some extracts from BCBS Standard relating to earnings are worth reminding as they clearly highlight that earnings are considered as NII:
§8. IRRBB refers to the current or prospective risk to the bank’s capital and earnings arising from adverse movements in interest rates that affect the bank’s banking book positions. When interest rates change, the present value and timing of future cash flows change. This in turn changes the underlying value of a bank’s assets, liabilities and off-balance sheet items and hence its economic value. Changes in interest rates also affect a bank’s earnings by altering interest rate-sensitive income and expenses, affecting its net interest income (NII). Excessive IRRBB can pose a significant threat to a bank’s current capital base and/or future earnings if not managed appropriately.
§69. The level of IRRBB exposure should be measured and disclosed. Specifically, banks must disclose the measured ΔEVE and ΔNII under the prescribed interest rate shock scenarios set out in Annex 2.
§93. A bank could also be considered to have excessive risk relative to earnings if its shocked ΔNII was such that the bank would not have sufficient income to maintain its normal business operations.
Implementation deadline:
• We would like to stress that the implementation of the Guideline will require a significant time to be carried out. Consequently, a sufficient period (at least 1 year) should be assigned before application. Ideally, the final application date of the Guideline should be aligned with the application of the technical standards, as the EBA suggested already in article 8.
Several Issues should be Fixed
The definition of interest rate sensitive instruments should be modified as it is inconsistent with the very definition of the BB and as it would leave balance sheet items out of both the Trading Book and out of the BB when combined to the definition of interest rate risk arising from non-trading book.
Hence, the definition below:
Interest rate sensitive instruments: Assets, liabilities and off-balance-sheet items in the non-trading book which are sensitive to interest rate changes (excluding assets deducted from CET1 capital, e.g. real estate or intangible assets or equity exposures in the non-trading book).
… should be substituted by:
Interest rate sensitive instruments: Interest rate risk arising from all items not included in the trading book.
Excluding items from both the trading book and the BB would lead to inconsistent measurement and flawed management. As an illustration, real estate or intangible assets create interest rate risk through their financing needs that are interest bearing.
The changes above would ensure consistency with §48(i): (i) Whether or not non-interest-bearing assets and liabilities of the non-trading book (including capital and reserves) are included in calculations measuring IRRBB for the ICAAP.
For sake of clarity, the definition of interest rate risk arising from non-trading book activities should also be modified:
Hence, the definition below:
Interest rate risk arising from non-trading book activities: The current or prospective risk to both the net interest income and the economic value of an institution arising from adverse movements in interest rates that affect interest rate sensitive instruments, including gap risk, basis risk and option risk.
… should be substituted by:
Interest rate risk arising from non-trading book activities: The current or prospective risk to both the net interest income and the economic value of an institution arising from adverse movements in interest rates that relates from the banking book (i.e. all items not included in the trading book).
The definition of credit spread sensitive instruments should be modified to be consistent with the definition of CSRBB:
Hence, the definition below:
Credit spread sensitive instruments: Assets, liabilities and off-balance-sheet items in the non-trading book, which are sensitive to credit spread changes (excluding assets deducted from CET1 capital, e.g. real estate or intangible assets or equity exposures in the non-trading book).
… should be substituted by (using EBA 2018 Guideline definition of CSRBB):
Credit spread sensitive instruments: Assets in the non-trading book, which are sensitive to changes in the market perception about the price of credit risk, liquidity premium and potentially other components of credit-risky instruments inducing fluctuations in the price of credit risk, liquidity premium and other potential components, which are not explained by IRRBB.
The definition of CSRBB should be clarified and simplified as it is difficult to understand and will lead to complex implementation:
Hence, the definition below:
Credit spread risk from non-trading book activities (CSRBB): The risk driven by changes of the market price for credit risk, for liquidity and for potentially other characteristics of credit-risky instruments, which is not captured by IRRBB or by expected credit/(jump-to-) default risk. CSRBB captures the risk of an instrument’s changing spread while assuming the same level of creditworthiness, i.e. how the credit spread is moving within a certain rating/PD range.
… should be substituted by (using EBA 2018 Guideline definition of CSRBB):
Credit spread risk from non-trading book activities (CSRBB): The risk driven by changes in the market perception about the price of credit risk, liquidity premium and potentially other components of credit-risky instruments inducing fluctuations in the price of credit risk, liquidity premium and other potential components, which are not explained by IRRBB.
The definition of Net Interest Income Measures should be corrected as it deviates from Net Interest Income:
Hence, the definition below:
Measures of changes in expected future profitability within a given time horizon resulting from interest rate movements, in case of IRRBB; or from credit spread changes, in case of CSRBB. It encompasses interest income, interest expenses and market value changes.
… should be substituted by:
Measures of changes in expected future interest income or interest expenses resulting from interest rate movements, in case of IRRBB; or from credit spread movements, in case of CSRBB.
The definition of Economic Value (EV) measures should be modified to be consistent with CSRBB that is defined in relation to market value, not net present value:
Hence, the definition below:
Measures of changes in the net present value of interest rate sensitive instruments over their remaining life resulting from interest rate movements, in case of IRRBB; or of changes in the net present value of instruments sensitive to credit spread changes over their remaining life resulting from credit spread movement, in case of CSRBB. EV measures reflect changes in value over the remaining life of the interest rate sensitive instruments, in case of IRRBB, or of the credit spread risk sensitive instruments, in case of CSRBB, i.e. until all positions have run off.
… should be substituted by:
Measures of changes in the net present value of interest rate sensitive instruments over their remaining life resulting from interest rate movements in case of IRRBB; or of changes in the market value of instruments sensitive to credit spread changes in case of CSRBB. EV measures reflect changes in value over the remaining life of the interest rate sensitive instruments, in case of IRRBB, or of the credit spread risk sensitive instruments, in case of CSRBB, i.e. until all positions have run off.
The definition of Economic Value of Equity (EVE) Measures should be clarified as equity is not defined:
Hence, the definition below:
A specific form of EV measure where equity is excluded from the cash flows.
… should be substituted by:
A specific form of EV measure for interest rate where CET1 items are excluded from the cash flows.
There is no need to define Retail or Transactional deposits and accounts in this Guideline that should not be prescriptive on details of IRRBB management framework. In addition, those notions are already defined in liquidity regulatory framework.
The reference to going concern is confusing in §16 that should be amended from:
§16 The change in the net interest income should be the difference between the expected net interest income under a shock or stress scenario from a going-concern perspective and the expected net interest income under a base scenario
… should be substituted by:
§16 The change in the net interest income should be the difference between the expected net interest income under a shock or stress scenario and the expected net interest income under a base scenario
Though risk appetite should factor in all significant risks, they should not necessarily be considered in sub-type in isolation, even less so as those subtypes originate from the very same drivers (interest rate) and transactions. Hence, §34 should be amended:
§34. The institution’s risk appetite for IRRBB should be expressed in terms of the acceptable impact of fluctuating interest rates on both net interest income and economic value and should be reflected in limits. Institutions with significant exposures to gap risk, basis risk or option risk should determine their risk appetite in relation to each of these material sub-types of IRRBB.
… should be substituted by:
§34. The institution’s risk appetite for IRRBB should be expressed in terms of the acceptable impact of fluctuating interest rates on both net interest income and economic value and should be reflected in limits. Identified significant exposures to gap risk, basis risk or option risk should be factored in their risk appetite.
As §41(a) clarifies that the management body bears ultimate responsibility but may delegate the monitoring and management of IRRBB to senior management, expert individuals or an asset and liability management committee under the conditions further specified in §43, it should be reflected in the rest of the guideline (as in §51, §65, §69, §75, §101 for instance The management body or its delegates):
§42. The management body, or its delegates, should […]
§47. The management body, or its delegates, should […]
There is no rational in isolating specific items in the banking book when they are managed together as a net exposure. Hence, paragraphs 68(e) and (f) should be deleted:
As commercial margins and other spread components may be be excluded from the economic value measures, which is consistent to focus on the interest rate risk, it would be consistent to apply the same approach for net interest income measures.
Articles §82 and §83 should be modified from:
§82. If commercial margins and other spread components are excluded from economic value measures, institutions should (i) use a transparent methodology for identifying the risk-free interest rate at inception of each instrument; and (ii) use a methodology that is applied consistently across all interest rate sensitive instruments and all business units.
§83. When calculating net interest income measures to evaluate IRRBB exposures, institutions should include commercial margins.
… to:
§82. If commercial margins and other spread components are excluded from economic value or net interest income measures, institutions should (i) use a transparent methodology for identifying the risk-free interest rate at inception of each instrument; and (ii) use a methodology that is applied consistently across all interest rate sensitive instruments and all business units.
§83. deleted
The reverse stress test should be done within the enterprise wide stress tests whereby the other risk categories (credit risk, market risk…) are considered. Considered in isolation, IRRBB does not lend itself to reverse stress test. We recommend modifying §99:
§99. In enterprise-wide stress tests, Institutions should perform reverse stress tests in order to (i) identify interest rate scenarios that could severely threaten an institution’s capital, economic value and net interest income; and (ii) reveal vulnerabilities arising from its hedging strategies and the potential behavioural reactions of its customers.
As described above, there is no direction for ‘prudence’ (notably shorter is not more prudent than longer). Hence, paragraph 110(c) and 112(e) should be amended like this:
§110(c) Ensure that modelling of key behavioural assumptions is justifiable in relation to the underlying historical data, and based on sound hypotheses;
§112(e) Ensure that assumptions about the decay of core and other modelled balances are sound and appropriate in balancing the benefits to net interest incomes against the additional economic value risk entailed in locking in a future interest rate return on the assets financed by these balances, and the potential forgone revenue under a rising interest rate environment.
§113(b) Determine
the elected investment maturity profile for the eligible equity capital that balances the benefits of income stabilisation arising from taking longer-dated fixed-return positions against the additional economic value sensitivity of those positions under an interest rate stress, and the risk of net interest income underperformance should rates rise;
Question 1: In the context of the measurement of the impact of IRRBB under internal systems, paragraph 111 envisages a five year cap repricing maturity for retail and non-financial wholesale deposits without a specified maturity. Would you foresee any unintended consequence or undesirable effect from this behavioural assumption in particular on certain business models or specific activities? If this is the case, please kindly provide concrete examples of it.
Executive Summary:The French Banking Federation (FBF) welcomes the opportunity to express the views of the French banking industry on the public consultation on the draft Guidelines on the identification, evaluation, management and mitigation of risks arising from potential changes in Interest Rates and of the assessment and monitoring of Credit Spread risk, of institutions’ non-Trading Book activities. In this context, we herewith provide you with our general remarks and responses to the questions listed in the Consultation Paper (CP). We appreciate your consideration about our comments and remain at your disposal for further clarifications.
Article 84(6) of Capital Requirement Directive (CRD) tasks the European Banking Authority (EBA) with developing a Guideline to define:
• the evaluation by an institution's internal system;
• the identification, management and mitigation by institutions of the risks;
• the assessment and monitoring by institutions of the risks;
• determining which of the internal systems are not satisfactory.
The European Banking Authority (EBA) has published Guidelines on Interest Rate Risk in the Banking Book (IRRBB) in May 2015 and July 2018 and is to publish yet another version of its Guideline on the very same topic in the course of 2022. There is a need to a stable regulatory framework and we are hopeful that it will be stabilized.
In July 2018, EBA elected to modify its May 2015 Guideline to factor in Basel Committee on Banking Supervision (BCBS) Standard on IRRBB published in April 2016 ahead of the mandate to be received from CRR to factor in this very same BCBS Standard. Hence, the EBA 2018 Guideline should be only marginally modified.
However, the Consultation Paper (CP) introduces very significant deviations from July 2018 Guideline that are not substantiated by any change in BCBS or CRR mandate and actually can be considered as:
• the CP envisages to change the definition of commonly understood Net Interest Income (NII) to include changes in fair values of instruments even though they are not part of NII. This appears as a deviation from the CRR mandate that explicitly refers to NII. It also deviates from BCBS Standards that is quite explicit that it refers to NII excluding changes in fair values that don’t affect NII. This would also be at odds with actual risk management and would introduce overlapping between NII measures and Economic Value (EV) measures while they should be complementary. NII should be kept as defined by interest income and expenses.
• The CP envisages a new limitation to the duration of Non-Maturing Deposits (NMD’s) (i.e. a 5 year cap on NMD’s from retail and non-financial wholesale, and a 0 year cap on NMD’s from financial customers) following a said “prudent” approach. These caps are inconsistent with a prudent approach since interest rate risk is a symmetric risk. Without any economic justification, they would lead to inappropriate measurement and management if incorporated in a management framework. It is reminded that, such caps are neither in CRR/CRD nor in BCBS Standard (where such caps appear only in the non-mandatory standardized method). Furthermore, the mandate to the EBA on the definition of SOT explicitly excludes behavioral assumption (CRD.Art.98(5a): ‘common modelling and parametric assumptions, excluding behavioral assumptions), the introduction of caps on NMD’s in the general framework of IRRBB management appears as EBA working around the mandates that it received.
Caps on NMD’s should be deleted.
• the CP envisages dramatic changes to the definition and scope of Credit Spread Risk in the Banking Book (CSRBB) while the July 2018 EBA Guideline already implemented the BCBS Standard that has not changed since then. The envisaged changes are not only not substantiated but they would also introduce significant confusions and complexities. CSRBB should be unchanged compared to July 2018 EBA Guideline, notably with a scope defined by assets that are actively traded on a deep and liquid markets to have identifiable and measurable market perception and changes thereof.
CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation are limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
• the CP provides principles for the assessment by competent authorities of the management framework for IRRBB and CSRBB. Broadly, those principles make sense to cover the different dimensions of the management framework. However, to the extent that the assessment could lead competent authorities to require banks to apply the Standardized Methodology for the evaluation of IRRBB, it is essential to better frame the conditions under which banks such a requirement could apply.
The requirement to apply the Standardized Methodology for the evaluation should relate to the evaluation component of the Internal Management System (IMS) that should be evidenced as non-satisfactory. In other words, a framework that would be assessed as non-satisfactory for issues that are not related to its evaluation component of its IMS should not lead to require using the Standardized Methodology.
As any Standardized Methodology would not be risk sensitive, as made clear by BCBS, it should be clarified that requiring its application should follow a careful analysis to evidence whether the Standardized Methodology would be more risk sensitive than the IMS evaluation it would substitute. Bar such an evidence, Standardized Methodology should not apply. Moreover, the GL should require competent authorities to grant institutions the possibility to remedy potential shortcomings in their IMS within a reasonable period of time before the Standardized Methodology is imposed. Last but not least, if the Standardized Methodology was imposed after all, this should only be temporary and a return to IMS should be allowed as soon as possible.
• We would like to stress that the implementation of the Guideline will require a significant time to be carried out. Consequently, a sufficient period (at least 1 year) should be assigned before application. Ideally, the final application date of the Guideline should be aligned with the application of the technical standards, as the EBA suggested already in article 8.
There are many other necessary changes to the draft Guideline to make it appropriate and consistent, notably on the suggested definitions (many of which changes since July 2018 EBA Guideline!) and the notion of ‘prudence’. Interest rate risk is a symmetrical risk: it is as risky to adopt too short durations than too long durations when considering banking book items. The CP is biased in considering prudence as leaning towards shorter duration, which does not make economic sense.
The changes are presented in detail below.
We are committed to elaborate a sound and robust management framework for IRRBB and CSRBB and remain available to EBA to support its efforts.
Response to Question 1 :
• The CP envisages a new limitation to the duration of Non-Maturing Deposits (NMD’s) (i.e. a 5 year cap on NMD’s from retail and non-financial wholesale, and a 0 year cap on NMD’s from financial customers) following a said “prudent” approach. These caps are inconsistent with a prudent approach since interest rate risk is a symmetric risk. Without any economic justification, they would lead to inappropriate measurement and management if incorporated in a management framework. It is reminded that, such caps are neither in CRR / CRD nor in BCBS Standard (where such caps appear only in the non-mandatory standardized method). Furthermore, the mandate to the EBA on the definition of SOT explicitly excludes behavioral assumption (CRD.Art.98(5a): ‘common modelling and parametric assumptions, excluding behavioral assumptions), the introduction of caps on NMD’s in the general framework of IRRBB management appears as EBA working around the mandates that it received.
Caps on NMD’s should be deleted.
There should be no limitation to the development of behavioral models that are instrumental in the risk-sensitive management of IRRBB. In particular, there should be no limitation to the maturity of deposits.
All of the customer-related Banking Book (BB) items require behavioral models to convey the actual interest rate risk exposure. Such models are essential to factor in the idiosyncrasies of each product, product mix, customer, competition, business model (including customer relationship, product mix), regulation, law and tax environment. Those idiosyncrasies need to be considered in the measurement, monitoring and management of IRRBB.
This has been recognized by BCBS in its Standard published as recently as 2016 (confirming the previous 2004 BCBS Standard) and confirmed in the CRD V Article 98.5a which clearly excludes from the EBA mandate the possibility to specify assumption in relation with behavioral assumptions for SOT calculus. As a consequence, the proposed five-year cap should be considered as both out of the mandate of EBA and detrimental to the quality of the IRRBB management framework of European banking institutions.
The statement that the EBA proposal is ‘less strict than the Basel standardized approach’ (p41 of the explanatory box of CP 2021/37) is inconsistent and misleading as it wrongly compares the management framework based on internal models that the BCBS fully recognizes without limitation to NMD’s. Actually, the BCBS Standard only suggests a standardized approach limited to be a fallback approach in case of non-satisfactory internal approach and in this fallback approach, NMD’s are clearly described as non-amenable to standardization. The introduction of any binding hard-coded limitation in the internal models would actually be a deviation from the BCBS Standard.
Any binding hard-coded limitation to the behavioral models would lead to failing to represent the actual interest rate risk exposure, would lose the fundamental risk-sensitivity of the IRRBB measurements and result in mitigating transactions that could actually magnify the interest rate risk exposure rather than mitigate it. It hardly can be a regulatory or supervisory objective to impose a fundamentally flawed measurement, monitoring and management of interest rate risk that it is supposed to regulate / supervise. In such binding hard-coded limitation would be a fatal flaw in the regulatory / supervisory framework.
For instance, the natural offsets with other interest rate risk exposures in the BB (e.g. fixed interest rate loans, notably mortgages) would not be recognized, leading to apparent and wrong interest rate risk exposure. This would lead to entering into mitigation derivative transactions to ‘hedge’ those flawed interest rate risk exposures that would actually increase the interest rate risk exposure. This could also lead banks to limit the extension of long-term fixed rate loans that their customers demand, exposing customers to changes in interest rate risk (as they will be exposed to increase in interest rates when repricing / rolling over their loans). None of this makes sense.
As an illustration, French customers demand overwhelmingly fixed interest rate mortgages that are naturally offset by a significant portion of French customers non-interest-bearing demand deposits whose stability has been evidenced over very long-term period and through very different interest rate environment.
This also applies to specific products such as Plan d’Epargne Logement (PEL) which is a long dated fixed rate savings accounts and to regulated savings accounts (e.g. Livret A) which exist since the 19th century.
Some extracts from BCBS Standard relating to NMD’s are worth reminding, while reminding that BCBS does not prescribe any cap on NMD’s neither in the management framework recommended by BCBS, nor in the SOT (Principle 10):
§3. The Committee noted the industry’s feedback on the feasibility of a Pillar 1 approach to IRRBB, in particular the complexities involved in formulating a standardised measure of IRRBB which would be both sufficiently accurate and risk-sensitive to allow it to act as a means of setting regulatory capital requirements. The Committee concludes that the heterogeneous nature of IRRBB would be more appropriately captured in Pillar 2.
Process for positions not amenable to standardisation […] §109. Positions not amenable to standardisation include (i) NMDs […]
The perception that it is prudent to be shorter than longer does not make any economic or supervisory sense. Interest rate risk is a symmetrical risk: being too short is as risky as being too long.
For these reasons, we urge EBA to not impose any such hard-coded limitation to behavioral model and to delete paragraph 111.
Indeed, the paragraph below should be deleted:
§111. The assumed behavioural repricing date for retail and non-financial wholesale deposits without any specific repricing dates (non-maturity deposits) should be constrained to a maximum weighted average repricing date of 5 years. The 5-year cap applies individually for each currency. Non-maturity deposits from financial customers should not be subject to behavioural modelling.
Question 2: Do respondents find that the criteria to identify non-satisfactory IRRBB internal models provide the minimum elements for supervisors’ assessment?
The paragraph §119 of the EBA CP defines the case when internal systems should be considered as not satisfactory:More specifically, internal systems should be considered as not satisfactory in the following cases at least:
a) An IMS should be considered non-satisfactory for the purposes of paragraph 3 of Article 84 CRD if competent authorities assess, on a case by case basis, that the implemented methods do not cover all the material components of the interest rate risk (gap risk, basis risk, option risk), and/or measures do not capture in a robust and economically justified manner all material dimensions of risks for significant assets, liabilities and off-balance sheet type instruments (e.g. NMD, loans, options) of the bank’s non-trading book.
Annex I describes a non-restrictive list of methods for IRRBB measurement with an indication of their limitations.
When measuring their exposure to IRRBB, institutions should not limit themselves to the methods listed in Annex I in order to ensure that material aspects of interest rate risk are captured adequately.
b) IMS should be considered as non-satisfactory if they are not calibrated, back tested and reviewed in all their relevant parameters on an appropriate frequency and supported by a due governance and documentation that considers the nature, scale and complexity of the IRRBB inherent in the business model and the institution's activities.
Institutions should duly comply with para. 71 to 79 of these Guidelines in particular with review and validation (at their appropriate frequency, including back testing), governance, risk policies as well as controls.
Broadly, those statements make sense. However, experience shows that supervisor may apply some of EBA Guideline at the letter without applying proportionality principle. As the consequences of being considered non-satisfactory are important, notably with the use of the Standardized Approach that will fail to accurately represent interest rate risk exposure and will lead to wrong interest rate risk management decision, an internal system should be assessed as non-satisfactory only when the non-satisfactory points are evidenced as being material for measurement and management.
The paragraph 119(b) appears too detailed as the modeling framework is already referred to §71-79,. It is also potentially misleading as it may be understood as requiring each single parameter to be subject to back-testing rather than models as a whole, and as some assumptions are necessarily discretionary (e.g. choice of the duration of equity).
We propose amendments to §119 like below:
More specifically, considering the nature, scale and complexity of the IRRBB inherent in the business model and the institution's activities and applying proportionality principle in the assessment, internal systems should be considered as not satisfactory in the following cases:
a) An IMS should be considered non-satisfactory for the purposes of paragraph 3 of Article 84 CRD if competent authorities assess, on a case by case basis, that the implemented methods do not cover all the material components of the interest rate risk (gap risk, basis risk, option risk), and/or measures do not capture in a robust and economically justified manner all material dimensions of risks for significant assets, liabilities and off-balance sheet type instruments (e.g. NMD, loans, options) of the bank’s non-trading book.
Annex I describes a non-restrictive list of methods for IRRBB measurement with an indication of their limitations.
When measuring their exposure to IRRBB, institutions should not limit themselves to the methods listed in Annex I in order to ensure that material aspects of interest rate risk are captured adequately.
b) IMS should be considered as non-satisfactory if they lead to material deviation from the application of para. 71 to 79 of these Guidelines in particular with review and validation, governance, risk policies as well as controls.
In addition, it should be made clear that the requirement to apply the Standardized Methodology for the evaluation should relate to the evaluation component of the Internal Management System (IMS) that should be evidenced as non-satisfactory. In other words, a framework that would be assessed as non-satisfactory for issues that are not related to its evaluation component of its IMS should not lead to require using the Standardized Methodology.
As any Standardized Methodology would not be risk sensitive, as made clear by BCBS, it should be clarified that requiring its application should follow a careful analysis to evidence whether the Standardized Methodology would be more risk sensitive than the IMS evaluation it would substitute. Bar such an evidence, Standardized Methodology should not apply. Moreover, the GL should require competent authorities to grant institutions the possibility to remedy potential shortcomings in their IMS within a reasonable period of time before the Standardized Methodology is imposed. Last but not least, if the Standardized Methodology was imposed after all, this should only be temporary and a return to IMS should be allowed as soon as possible.
Question 3: Is there any specific element in the definition of CSRBB that is not clear enough for the required assessment and monitoring of CSRBB by institutions?
• The CP envisages dramatic changes to the definition and scope of Credit Spread Risk in the Banking Book (CSRBB) while the July 2018 EBA Guideline already implemented the BCBS Standard that has not changed since then. The envisaged changes are not only not substantiated but they would also introduce significant confusions and complexities. CSRBB should be unchanged compared to July 2018 EBA Guideline, notably with a scope defined by assets that are actively traded on a deep and liquid markets to have identifiable and measurable market perception and changes thereof.CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation is limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
We consider that CSRBB guidelines are quite unclear. More specifically, the different spread components definitions are not so clear and the practical procedure to identify them neither.
See Response to Question #5 for more details.
Question 4: As to the suggested perimeter of items exposed to CSRBB, would you consider any specific conceptual or operational challenge to implement it?
• The CP envisages dramatic changes to the definition and scope of Credit Spread Risk in the Banking Book (CSRBB) while the July 2018 EBA Guideline already implemented the BCBS Standard that has not changed since then. The envisaged changes are not only not substantiated but they would also introduce significant confusions and complexities. CSRBB should be unchanged compared to July 2018 EBA Guideline, notably with a scope defined by assets that are actively traded on a deep and liquid markets to have identifiable and measurable market perception and changes thereof.CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation is limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
The proposed perimeter is much larger than the one currently considered by European and French banks. Furthermore, the methodology to be applied is not clear for us.
Cf. Response to Question #5 for more details.
IRRBB only
See CSRBB only and Both IRRBB and CSRBBCSRBB only
CSRBB:• the CP envisages dramatic changes to the definition and scope of Credit Spread Risk in the Banking Book (CSRBB) while the July 2018 EBA Guideline already implemented the BCBS Standard that has not changed since then. The envisaged changes are not only not substantiated but they would also introduce significant confusions and complexities. CSRBB should be unchanged compared to July 2018 EBA Guideline, notably with a scope defined by assets that are actively traded on a deep and liquid markets to have identifiable and measurable market perception and changes thereof.
CSRBB should relate to fair-valued assets that are actively traded on a deep and liquid markets so as to have identifiable and measurable market perception and changes thereof. Derivatives, if any, that are hedging CSRBB should be clarified as included in CSRBB. Pension obligations and pension plan assets should be clarified as being excluded from CSRBB.
We also noted that the EBA by means of article 14 has expanded the scope of article 84.6c for CSRBB, where institutions are only obliged to assess and monitor credit spread risk. Article 14 of this Guideline would also oblige institutions to control credit spread risk. However, in article 84 the CRD explicitly differentiates between IRRBB and CSRBB, where the management and mitigation are limited to IRRBB. Therefore, asking institutions to control credit spread is beyond the mandate of the EBA and not deemed proportionate.
As EBA elected to factor the BCBS 2016 Standard in its 2018 EBA Guideline, ahead of the finalization of the CRR, and as there was no other elements published by BCBS, we expect EBA Guideline to remain unchanged compared to 2018 EBA Guideline with the definition of CSRBB:
Definition (p14): The risk driven by changes in the market perception about the price of credit risk, liquidity premium and potentially other components of credit-risky instruments inducing fluctuations in the price of credit risk, liquidity premium and other potential components, which is not explained by IRRBB or by expected credit/(jump-to-) default risk.
Scope (p17): Institutions should monitor and assess their CSRBB-affected exposures, by reference to the asset side of the non-trading book, where CSRBB is relevant for the risk profile of the institution.
In the CSRBB perimeter defined in §120, it is understood and agreed upon that CSRBB relates to items that are subject to a market perception, i.e. with sufficiently breadth and depth market. Hence, BB items that are not subject to a market, e.g. loans and deposits, are not part of CSRBB. This should be explicitly mentioned.
In §121, it is mentioned that ‘the deterioration of an institution’s credit quality should not have any positive impact on the credit spread risk measure’. As it would make no sense to consider that the improvement of an institution’s credit quality should be considered as a risk for the institution, it leads to the conclusion that credit spread risk can only relate to the asset side… as EBA concluded in its EBA 2018 Guideline.
We recommend that the scope of application of CSRBB is confirmed as covering the asset side of the non-trading book, as EBA 2018 Guideline and to mention explicitly that Loans and Deposits to be excluded from this scope.
Actually, CSRBB should be consistent with the business model in which those items are held. To the extent that those items, subject to a deep market, are not intended to be sold before their maturity, those items should typically not be considered in the scope of CSRBB. This applies to market-liquid debt securities held in a Collect mode (and consequently accounted for without any impact of their changes in market values). This is consistent with §152 (Institutions should implement robust internal measurement systems (IMSs) that capture all components and sources of CSRBB which are relevant for the institution’s business model.).
As hedging derivatives in the BB are usually collateralized (directly or indirectly by their internal trading department), and subject to counterparty risk framework, they would not be subject to CSRBB except if they mitigate CSRBB.
We recommend deleting the explanatory box ‘Examples of CSRBB identification and measurement’ as it is too prescriptive and too complex:
• economic value perspective does not mean necessarily Economic Value of Equity (EVE)
• the breakdown by rating may not be the most appropriate (e.g. breakdown by issuer type may be more relevant)
• the breakdown by terms appears too prescriptive and complex
As a conclusion, the EBA Guideline should be modified:
Paragraph §124 should be modified from:
§124. Institutions should not exclude any instrument in the banking book from the perimeter of CSRBB ex ante, including assets, liabilities, derivatives and other off-balance sheet items such as loan commitments, irrespective of their accounting treatment. Any potential exclusion of instruments from the relevant perimeter should be done in the case of the absence of sensitivity to credit spread risk and should be appropriately documented and justified. In any case, institutions should not exclude assets accounted at fair value.
… to:
§124. Institutions should include the banking book assets that are actively traded on a deep and active markets (i.e. Level 1 assets) in the perimeter of CSRBB held in a collect and sell a business model. Any potential exclusion of instruments from the relevant perimeter should be done in the case of the absence of sensitivity to credit spread risk and should be appropriately documented and justified. In any case, institutions should not exclude assets accounted at fair value.
In terms of governance, the management body should be able to delegate its responsibility for CSRBB as for IRRBB:
Paragraph §130(a) should be aligned on §41(a):
§130(a) That their management body bears the ultimate responsibility for the oversight of the CSRBB management framework and the institution’s risk appetite framework to adequately cover the risks. The management body may, however, delegate the monitoring and management of IRRBB to senior management, expert individuals or an asset and liability management committee under the conditions further specified in paragraph 132.
Then, the management body should be systematically accompanied with ‘or its delegates’ (as in §131(e), §135, §136, §145, §146, §148, §149, §151):
§131 The management body or its delegates […]
§131(d) […] The level of and changes ni the institution’s CSRBB exposure should be provided regularly to the management body or its delegates.
§131(f) Understanding and assessing the functioning of its delegates in monitoring and controlling CSRBB, consistent with policies approved by the management body or its delegates, on the basis of regular reviews of timely and sufficiently detailed information.
§133 The management body or its delegates […]
The articulation and separation between CSRBB and IRRBB need to be clarified as it was in the EBA 2018 Guideline: it should be clear that CSRBB excludes IRRBB.
The exclusion of idiosyncratic components would make CSRBB far more complex as market perceptions (ability to retrieve prices from an organized liquid market) do not segregate the idiosyncratic component. Isolating it would be theoretical as methods would need to be developed with introducing models. The discretion that EBA CP envisages in §157 to include the idiosyncratic component subject to being more conservative, would make it even more complex (§157: As an exception, in the practical implementation of paragraph 120 and for proportionality reasons, institutions may include idiosyncratic credit spread components for the monitoring of CSRBB, as long as it is ensured that the measures will yield more conservative results.) and would not ease practical implementation.
Moreover, with the implementation of the Benchmark Regulation, the market indicator that represented the liquidity premium of the AA banks that was the Libor will disappear for most major currencies with the current exception of EUR. It will no longer be possible to differentiate between purely idiosyncratic and systemic premiums.
We recommend that CSRBB is defined without exclusion of sub-components such as idiosyncratic.
Finally, we propose to delete §131(g) as it would apply to any single type of risk and there is no reason to state this specifically for CSRBB.
An extract from BCBS Standard relating to CSRBB are worth reminding as they clearly highlight that it relates to items that are at fair value (mark-to-market), as illustrated by the graph (re. red zone) below:
Annex 1 - §5 - Changes to the market liquidity spreads and market credit spreads are combined within the definition of CSRBB. The diagram below gives a visual representation of how the various elements fit together.
Both IRRBB and CSRBB
Net Interest Income should be… Net Interest Income as it is commonly understood• the CP envisages to change the definition of commonly understood Net Interest Income (NII) to include changes in fair values of instruments even though they are not part of NII. This appears as a deviation from the CRR mandate that explicitly refers to NII. It also deviates from BCBS Standards that is quite explicit that it refers to NII excluding changes in fair values that don’t affect NII. This would also be at odds with actual risk management and would introduce overlapping between NII measures and Economic Value (EV) measures while they should be complementary. NII should be kept as defined by interest income and expenses.
Net Interest Income (NII) means net interest income that is widely defined as the portion that impact the profit and loss statement.
Considering the changes in Other Comprehensive Income (OCI) and/or in capital as NII would be inconsistent with commonly accepted definition of NII, with CRD and with BCBS.
It would be so flawed as it would disincentivize to hedge with cash flow hedging instruments as their changes in fair value would be considered as a risk to NII while they are entered into precisely to make NII less sensitive.
We urge EBA to adhere to the common definition of NII, not to invent another definition of its own and to be consistent to the mandate provided by CRD that refers to Net Interest Income. EBA would deviate from its mandate by extending the definition of NII.
It is reminded that the economic value perspective has also to be considered and that the extension of NII measures would create overlap with EV measures.
Several paragraphs need to be fixed of this deviation like this:
15. For the purposes of these Guidelines, the net interest income upon which to calculate the impact of interest rate or credit spread movements should be determined by the interest income and expenses, and the market value changes of instruments — depending on accounting treatment in the profit and loss account. Institutions should take into account the increase or reduction in the amount of profit and losses over short- and medium-term horizons resulting from interest rate or credit spread movements.
§27(e) the impact on economic value and net interest income of mismatched positions in different currencies;
§31(d) The relative importance of interest rate sensitive instruments (including interest rate derivatives) in the non-trading book, with potential effects shown in the profit and loss account;
§44(c) In defining their risk appetites, institutions should take account of net interest income risks that may arise as a consequence of the accounting treatment of transactions in the non-trading book. The risk to net interest income may not be limited to interest income and expenses: the effects of changes in interest rates on the market value of instruments that, depending on accounting treatment, are reflected through the profit and loss account, should be taken into account. Institutions should particularly take into account the impact related to embedded optionalities in fair value instruments under ongoing interest rate shocks and stress scenarios. Institutions should also take into account the potential impact on the P&L accounts of hedging interest rate derivatives if their effectiveness was hampered by interest rate changes.
Some extracts from BCBS Standard relating to earnings are worth reminding as they clearly highlight that earnings are considered as NII:
§8. IRRBB refers to the current or prospective risk to the bank’s capital and earnings arising from adverse movements in interest rates that affect the bank’s banking book positions. When interest rates change, the present value and timing of future cash flows change. This in turn changes the underlying value of a bank’s assets, liabilities and off-balance sheet items and hence its economic value. Changes in interest rates also affect a bank’s earnings by altering interest rate-sensitive income and expenses, affecting its net interest income (NII). Excessive IRRBB can pose a significant threat to a bank’s current capital base and/or future earnings if not managed appropriately.
§69. The level of IRRBB exposure should be measured and disclosed. Specifically, banks must disclose the measured ΔEVE and ΔNII under the prescribed interest rate shock scenarios set out in Annex 2.
§93. A bank could also be considered to have excessive risk relative to earnings if its shocked ΔNII was such that the bank would not have sufficient income to maintain its normal business operations.
Implementation deadline:
• We would like to stress that the implementation of the Guideline will require a significant time to be carried out. Consequently, a sufficient period (at least 1 year) should be assigned before application. Ideally, the final application date of the Guideline should be aligned with the application of the technical standards, as the EBA suggested already in article 8.
Several Issues should be Fixed
The definition of interest rate sensitive instruments should be modified as it is inconsistent with the very definition of the BB and as it would leave balance sheet items out of both the Trading Book and out of the BB when combined to the definition of interest rate risk arising from non-trading book.
Hence, the definition below:
Interest rate sensitive instruments: Assets, liabilities and off-balance-sheet items in the non-trading book which are sensitive to interest rate changes (excluding assets deducted from CET1 capital, e.g. real estate or intangible assets or equity exposures in the non-trading book).
… should be substituted by:
Interest rate sensitive instruments: Interest rate risk arising from all items not included in the trading book.
Excluding items from both the trading book and the BB would lead to inconsistent measurement and flawed management. As an illustration, real estate or intangible assets create interest rate risk through their financing needs that are interest bearing.
The changes above would ensure consistency with §48(i): (i) Whether or not non-interest-bearing assets and liabilities of the non-trading book (including capital and reserves) are included in calculations measuring IRRBB for the ICAAP.
For sake of clarity, the definition of interest rate risk arising from non-trading book activities should also be modified:
Hence, the definition below:
Interest rate risk arising from non-trading book activities: The current or prospective risk to both the net interest income and the economic value of an institution arising from adverse movements in interest rates that affect interest rate sensitive instruments, including gap risk, basis risk and option risk.
… should be substituted by:
Interest rate risk arising from non-trading book activities: The current or prospective risk to both the net interest income and the economic value of an institution arising from adverse movements in interest rates that relates from the banking book (i.e. all items not included in the trading book).
The definition of credit spread sensitive instruments should be modified to be consistent with the definition of CSRBB:
Hence, the definition below:
Credit spread sensitive instruments: Assets, liabilities and off-balance-sheet items in the non-trading book, which are sensitive to credit spread changes (excluding assets deducted from CET1 capital, e.g. real estate or intangible assets or equity exposures in the non-trading book).
… should be substituted by (using EBA 2018 Guideline definition of CSRBB):
Credit spread sensitive instruments: Assets in the non-trading book, which are sensitive to changes in the market perception about the price of credit risk, liquidity premium and potentially other components of credit-risky instruments inducing fluctuations in the price of credit risk, liquidity premium and other potential components, which are not explained by IRRBB.
The definition of CSRBB should be clarified and simplified as it is difficult to understand and will lead to complex implementation:
Hence, the definition below:
Credit spread risk from non-trading book activities (CSRBB): The risk driven by changes of the market price for credit risk, for liquidity and for potentially other characteristics of credit-risky instruments, which is not captured by IRRBB or by expected credit/(jump-to-) default risk. CSRBB captures the risk of an instrument’s changing spread while assuming the same level of creditworthiness, i.e. how the credit spread is moving within a certain rating/PD range.
… should be substituted by (using EBA 2018 Guideline definition of CSRBB):
Credit spread risk from non-trading book activities (CSRBB): The risk driven by changes in the market perception about the price of credit risk, liquidity premium and potentially other components of credit-risky instruments inducing fluctuations in the price of credit risk, liquidity premium and other potential components, which are not explained by IRRBB.
The definition of Net Interest Income Measures should be corrected as it deviates from Net Interest Income:
Hence, the definition below:
Measures of changes in expected future profitability within a given time horizon resulting from interest rate movements, in case of IRRBB; or from credit spread changes, in case of CSRBB. It encompasses interest income, interest expenses and market value changes.
… should be substituted by:
Measures of changes in expected future interest income or interest expenses resulting from interest rate movements, in case of IRRBB; or from credit spread movements, in case of CSRBB.
The definition of Economic Value (EV) measures should be modified to be consistent with CSRBB that is defined in relation to market value, not net present value:
Hence, the definition below:
Measures of changes in the net present value of interest rate sensitive instruments over their remaining life resulting from interest rate movements, in case of IRRBB; or of changes in the net present value of instruments sensitive to credit spread changes over their remaining life resulting from credit spread movement, in case of CSRBB. EV measures reflect changes in value over the remaining life of the interest rate sensitive instruments, in case of IRRBB, or of the credit spread risk sensitive instruments, in case of CSRBB, i.e. until all positions have run off.
… should be substituted by:
Measures of changes in the net present value of interest rate sensitive instruments over their remaining life resulting from interest rate movements in case of IRRBB; or of changes in the market value of instruments sensitive to credit spread changes in case of CSRBB. EV measures reflect changes in value over the remaining life of the interest rate sensitive instruments, in case of IRRBB, or of the credit spread risk sensitive instruments, in case of CSRBB, i.e. until all positions have run off.
The definition of Economic Value of Equity (EVE) Measures should be clarified as equity is not defined:
Hence, the definition below:
A specific form of EV measure where equity is excluded from the cash flows.
… should be substituted by:
A specific form of EV measure for interest rate where CET1 items are excluded from the cash flows.
There is no need to define Retail or Transactional deposits and accounts in this Guideline that should not be prescriptive on details of IRRBB management framework. In addition, those notions are already defined in liquidity regulatory framework.
The reference to going concern is confusing in §16 that should be amended from:
§16 The change in the net interest income should be the difference between the expected net interest income under a shock or stress scenario from a going-concern perspective and the expected net interest income under a base scenario
… should be substituted by:
§16 The change in the net interest income should be the difference between the expected net interest income under a shock or stress scenario and the expected net interest income under a base scenario
Though risk appetite should factor in all significant risks, they should not necessarily be considered in sub-type in isolation, even less so as those subtypes originate from the very same drivers (interest rate) and transactions. Hence, §34 should be amended:
§34. The institution’s risk appetite for IRRBB should be expressed in terms of the acceptable impact of fluctuating interest rates on both net interest income and economic value and should be reflected in limits. Institutions with significant exposures to gap risk, basis risk or option risk should determine their risk appetite in relation to each of these material sub-types of IRRBB.
… should be substituted by:
§34. The institution’s risk appetite for IRRBB should be expressed in terms of the acceptable impact of fluctuating interest rates on both net interest income and economic value and should be reflected in limits. Identified significant exposures to gap risk, basis risk or option risk should be factored in their risk appetite.
As §41(a) clarifies that the management body bears ultimate responsibility but may delegate the monitoring and management of IRRBB to senior management, expert individuals or an asset and liability management committee under the conditions further specified in §43, it should be reflected in the rest of the guideline (as in §51, §65, §69, §75, §101 for instance The management body or its delegates):
§42. The management body, or its delegates, should […]
§47. The management body, or its delegates, should […]
There is no rational in isolating specific items in the banking book when they are managed together as a net exposure. Hence, paragraphs 68(e) and (f) should be deleted:
As commercial margins and other spread components may be be excluded from the economic value measures, which is consistent to focus on the interest rate risk, it would be consistent to apply the same approach for net interest income measures.
Articles §82 and §83 should be modified from:
§82. If commercial margins and other spread components are excluded from economic value measures, institutions should (i) use a transparent methodology for identifying the risk-free interest rate at inception of each instrument; and (ii) use a methodology that is applied consistently across all interest rate sensitive instruments and all business units.
§83. When calculating net interest income measures to evaluate IRRBB exposures, institutions should include commercial margins.
… to:
§82. If commercial margins and other spread components are excluded from economic value or net interest income measures, institutions should (i) use a transparent methodology for identifying the risk-free interest rate at inception of each instrument; and (ii) use a methodology that is applied consistently across all interest rate sensitive instruments and all business units.
§83. deleted
The reverse stress test should be done within the enterprise wide stress tests whereby the other risk categories (credit risk, market risk…) are considered. Considered in isolation, IRRBB does not lend itself to reverse stress test. We recommend modifying §99:
§99. In enterprise-wide stress tests, Institutions should perform reverse stress tests in order to (i) identify interest rate scenarios that could severely threaten an institution’s capital, economic value and net interest income; and (ii) reveal vulnerabilities arising from its hedging strategies and the potential behavioural reactions of its customers.
As described above, there is no direction for ‘prudence’ (notably shorter is not more prudent than longer). Hence, paragraph 110(c) and 112(e) should be amended like this:
§110(c) Ensure that modelling of key behavioural assumptions is justifiable in relation to the underlying historical data, and based on sound hypotheses;
§112(e) Ensure that assumptions about the decay of core and other modelled balances are sound and appropriate in balancing the benefits to net interest incomes against the additional economic value risk entailed in locking in a future interest rate return on the assets financed by these balances, and the potential forgone revenue under a rising interest rate environment.
§113(b) Determine
the elected investment maturity profile for the eligible equity capital that balances the benefits of income stabilisation arising from taking longer-dated fixed-return positions against the additional economic value sensitivity of those positions under an interest rate stress, and the risk of net interest income underperformance should rates rise;