Risposta di AIFIRM al consultative document “Interest rate risk in the banking book” del Comitato di Basilea sulla vigilanza bancaria

AIFIRM believes that the a Pillar 2 approach, where banks are allowed, subject to supervisory approval, to use internal measurement systems (IMS) for assessing their ability to cover potential losses from IRRBB, is the best suited option. In fact, the standardization associated with a Pillar 1 approach would lead to a lower precision of the risk exposure estimate and a poorer comprehension of the factors that determine it. By adopting a Pillar 1 approach, there is a higher probability that banks set aside an amount of internal capital that either underestimates or overestimates its appropriate amount, entailing a potential threat to the overall banking stability and an unnecessary reduction in lending capacity, respectively. AIFIRM welcomes the consideration of multiple scenarios because it is a step forward in the comprehension of risk determinants. The new mathematical framework allows to obtain a measure of risk exposure that is more consistent with the level of interest rates observed on the evaluation date and, therefore, represents an important improvement, if compared with the current one, which is based on unrealistic duration coefficients. The new framework can also be easily integrated within banks’ IMS for the implementation of more sophisticated methodologies. The Association believes that option n. 4 is the best-suited among those proposed by the Committee to calculate minimum capital requirements because it allows the net interest profit (NIP) to reduce minimum capital requirements associated with the change to economic value of equity (EVE) and earnings. This approach is based on the presence of many positions characterized by locked-in margins, which will generate a positive interest income even when EVE is at its highest. Following this method, minimum capital requirements are more consistent with banks’ actual riskiness and banks’ credit supply is calibrated in a more appropriate way. Within a Pillar 2 approach, minimum capital requirements may be based on both: i) stressed scenarios of changes in the key-rates that are consistent with the six proposed scenarios, in terms of both shock magnitude and structure, and ii) scenarios obtained through banks’ internal measurement systems. However, AIFIRM believes that further discussions and analyses on the NIP calibration are necessary. As concerns the treatment of the items characterized by behavioral options, AIFIRM recognizes the utility of introducing some constraints in modelling non-maturity deposits (NMDs), even in cases of banks’ own internal representations, since they could contribute to the reduction of the model risk. However, they seem to be too conservative giving rise, even in the discretionary approach, to a unique representation of NMDs. According to AIFIRM, based on the analysis of historical data of the Italian banking market: i) the allocation of the repricing component of NMDs may be led by the interest rate pass-through that follows a change in the reference market rate; ii) the core component of NMDs should include not only the fraction of non-maturity deposits that are stable, but also the portion that reprice, with a certain sluggishness, when the reference market rate changes. Finally, as regards positions other than NMDs, the Association agrees with the choice to model their optionalities using a two step approach. Within the supported solution of an enhanced Pillar 2 capital framework, AIFIRM believes that public disclosure on a regular basis of a bank’s IRRBB risk profile, key measurement assumptions, qualitative and quantitative assessment of IRRBB levels and quantitative disclosure of IRRBB metrics, is crucial. Banks should describe in detail the qualitative information required in BCBS (2015) for disclosure purposes since these are issues of particular relevance in estimating banks' risk exposure. As for the quantitative information, if appropriate public disclosure is important, disclosure of standardized calculation could be misleading. By using the standardized calculation, the proposed Pillar 2 approach is no different from the proposed Pillar 1. Supporting a “true” Pillar II approach, AIFIRM believes that banks' internal measurement and management of IRRBB are those which ought to be disclosed. This document is organized as follows: paragraph 1 comments on the choice between a Pillar 1 and a Pillar 2 solutions; in section 2, we discuss some of the main issues associated with the interest rate scenario design; section 3 deals with the specification of minimum capital requirements; in section 4 we analyze the treatment of the positions with behavioral options; section 5 provides some comments on the disclosure requirements.