Building Societies Association

NA
The general provisions in Section 4.1 are clear, but deficient in one respect : paragraph 20 should be explicitly qualified by reference to the size and complexity of the institution, as well as its existing / prospective exposure to IRRBB. Not all the elements are really necessary for small, simple banks. We suspect that paragraphs 16 (NPLs) and 18 (CSRBB) are examples of elements that could be disapplied for small banks. (Moreover, we understand there may be some discrepancy in the definition of CSRBB between the Guidelines and the BCBS Standards.)
NA
Section 4.2 on capital identification is sufficiently clear, but again lacks any explicit recognition of the need for proportionality. The requirements in section 4.2 should therefore be qualified by reference to the size and complexity of the institution – probably best located in paragraph 25.
NA
Section 4.3 regarding governance is clear, and in several places already recognises the proportionality principle, but we think this needs to be more fully embedded throughout Section 4.3.
Section 4.4 on measurement is clear and comprehensive. There is already welcome recognition in several places of the principle of proportionality ( e.g. at paragraphs 87, 89 and 91 (a) ) but we think this should go further, especially in terms of numbers of scenarios. Fewer but relevant scenarios applied, and their results analysed and understood more thoroughly, may give a better result in terms of practical risk management than larger numbers of scenarios that will inevitably be used more superficially, as people fail to see the wood for the trees.
NA
NA
No, section 4.5 does not adequately reflect the principle of proportionality. The gestures in that direction – extra six months for implementation and possibly reduced frequency – are wholly inadequate. While we support the principle of the “supervisory outlier” test as a consistent, comparable measure, as proposed it is too complex for small, simple banks. The over-complex items include any disaggregation of spreads, the inclusion of interest payment as well as principal cashflows, and the use of multiple interest rate floors across the yield curve. We also think NPLs and pension items could be dealt with more simply. The proposed treatments may be suitable for larger banks, but the cost-benefit equation for small, simple banks is adverse.
NA
NA
We doubt some of the systems in use by small, simple banks can necessarily disaggregate margins in doing the required calculations, nor do we think it reasonable to require expenditure on this. We suggest that this is an unnecessary refinement.
The “lower bound” involves a multiple stepped floor that we fear will prove excessively problematic for many banks’ systems, but add little value. A single floor, set say at zero across all time buckets, would be sufficient.
NA
NA
Jeremy Palmer
B