British Bankers' Association

Some members are of the opinion that the end user would benefit from having overview through the use of OV1-B whilst other prefer the summary on a page offered by OV1-A.
We prefer implementing the CRR approach until Phase 2 is implemented as it provides appropriate granularity between key segments including sovereigns and SME. To our understanding other industry associations have leaned towards COREP approach and we see this as a good opportunity to further harmonise between reporting and disclosure in view of pipeline CRR2/CRDIV developments and Phase 2 to avoid differences in approach and we request that EBA explores this further
We have no objections to this approach
Whilst this information may have some benefit to users (although there is limited evidence of this), and whilst with material investment it may be feasible to achieve the breakdowns, in light of clear links to IFRS 9 and the fact that IFRS 9 is not yet implemented we (a) do not recommend it is implemented at this time, rather that (b) it is implemented at/or after IFRS 9 implementation (on implementation of IFRS 9 banks should be more able to draw out range of granular data enabling them to provide requested information in Pillar 3 reports as per EBA proposal).
Whilst it is possible that some investors may be interested in the break-down of sources of counterparty credit risk by approach/model, members have not seen a material or increased level of such requests.

As such we believe that table CCR1 should only present on an IMM basis as risk management by banks is undertaken on overall portfolio basis and also further breakdowns would not be meaningful due to their limited materiality.

We also note concerns over the feasibility of providing this information. IMM is based on the use of probabilistic scenarios for the evolution of risk factors which are combined with pricing model outputs to determine future potential exposures. It is not feasible to attribute scenario outputs to transaction type due to their probabilistic nature and there is no practical solution to creating a meaningful breakdown.

In addition, we have noted other issues with providing a breakdown of CCR exposures by transaction type which would require a number of assumptions to be made:

- Netting assumptions – collateral used to net exposures is typically applied on a portfolio level.

- Allocation of diversification benefit - diversification benefits arise across a portfolio and assumptions would be needed to attribute them to each transaction type to ensure the total CCR exposure reconciled to the separate components in the breakdown, which would lead to inaccurate reporting at transaction type level.
Although revenue/P&L and position/mark to market reporting would split credit derivatives at individual transaction/client level (and thus between internal hedging and client transactions), most banks manage market risk on an overall/net portfolio basis. We do not favour Pillar 3 disclosures requiring a split that does not reflect risk management and existing reporting, would not be useful, not allow easy reconciliation between this segment of Pillar 3 and actual market risk reported (given the split would remove impacts of netting, portfolio effect, adjustments, overall collateral position etc.) and would reduce the reliability of information provided to regulators and investors. We would also ask for further clarification from the EBA on what its intended outcome is so that further enhancements can be considered for phase 2 in 2019.
Whilst the proposed form does not create additional burden for banks, it stops short of being useful and understandable for users in current form. We therefore recommend that the template is amended to the one below and this will also align with COREP templates C18, C21 and C24 - see table in main document attached.
We agree that proposed disclosure of end of period and average values for market risk metrics could provide useful insights for the users.
Overall, we agree with the current proposed scope of application of these guidelines and also understand that the current guidelines do not change the scope of application of the CRR disclosure requirements. However, it seems that specific points in the guidelines may have a broader scope of application such as point 4.3 section C which seems to go beyond what is required by CRR article 13 for significant subsidiaries (including subsidiaries of material significance to their local market). Indeed, we suggest that clarification is provided that disclosure required by significant subsidiaries of EU parent institutions or parent holding companies, regardless of whether they are O-SII or not, continue to be legally defined only by article 13. Furthermore, regarding proportionality, we understand that implementation of the revised Pillar 3 Framework will eventually require a revision of the CRR in its part eight. This would seem a more appropriate route to better address the principle of proportionality.
We support the introduction of a key risk template as it will provide a high level overview and will facilitate comparison across banks key metrics. Indeed, we believe something similar to the BCBS proposal for a Key Metrics template (KM1) published within its Pillar 3 phase 2 disclosure requirements consultation would be appropriate and as such should follow after the Basel template is finalised.
Separate from anecdotal interest, we have not observed a large volume of actual requests received as per member feedback. Whilst we recognise an underlying and evolving user need for disclosures in editable format, such an approach would generate significant additional governance and revalidation burden, especially if implemented in the short term. This is disproportionate to observed actual demand and thus we do not support immediate/short term roll-out of reports in editable format and recommend appropriate proportionality is applied with a phased approach from Phase 2 which would start with the 11 core templates before progressing to remaining tables. This would assure that availability of information continues to evolve in a manner that allows implementation of appropriate resources and governance to provide assurances to investors as to the accuracy of information they are receiving.
Please refer to our response to Question 11 directly above.
We believe adoption of the 11 templates recommended for early implementation may be feasible. We understand that one of the purposes of these guidelines is to relieve the burden of having to prepare two sets of disclosures for European institutions that need to provide the revised Pillar 3 disclosures while still legally bound to comply with CRR requirements. The timeframe for implementation of the guidelines, however, is not appropriate as the EBA has said that the guidelines will not be finalised until the end of 2016. As a consequence, mandatory application of end-2016 disclosures should not be specified, although adoption could be recommended. It therefore should be appreciated that while it may be feasible to adopt the initial 11 templates, when considering further change it should be noted that not everything can be achieved in the same way. We also recommend that any early release of further templates on a trial basis before Q4 2017 is not treated by EBA or local regulators as full operationalisation and that the obligation to provide additional tables only becomes applicable from Q4 2017.

Furthermore, we recommend that a reasonable period of time is given to banks to produce these disclosures and that the “in conjunction with” financial statements reporting timeline is defined as eight weeks after period end, which would allow for sufficient time to carry out any reconciliations and internal governance before Pillar 3 disclosures are published. To our understanding, this aligns to views of other key European banks.
See comments on page 2 of the main document attached relating to CCR and MR and Appendix 1 for detailed observations.
We welcome the EBA initiative to provide more specific guidance and also introduce specific formats for disclosures that are in line with both BCBS proposals and CRR requirements.

We would like to emphasise the importance of consistency and comparability of institutions’ regulatory disclosures. EBA Guidelines should provide clarifications where required as banks following their own interpretations could lead to inconsistencies, which is contradictory to the purpose of these guidelines. In the detailed comments on the templates as provided below, we have pointed out various instances where further clarification is required so as to avoid inconsistent disclosures.
We find that the impact assessment did not factor in material personnel, IT and financial burdens that the new requirements would put on individual banks, both at launch/implementation (in view of the short period for roll-out) and on ongoing basis (in view of high frequency of reporting and proximity to key financial reporting periods). It is further our view that the impact assessment didn’t factor in the increased ongoing (and implementation) cost of additional:

- Reconciliation, transformation, aggregation and governance

- Reporting in excess of BCBS requirements or of implementing IFRS 9 related workarounds/partial solution prior to full implementation of IFRS 9 (which may lead to a further increase in cost for removal/change of solutions put in place to meet P3 timeline)

- Continuing duplication of EBA Pillar 3 and BCBS reporting for banks (and our members) that operate outside EU/EEA and who continue to be obliged to provide BCBS reports as well

- Additional cost to finance and IT departments of banks which are facing parallel roll-out of number of regulatory changes impacting similar processes and systems but with unaligned requests (IFRS 9, EBA P3, BCBS, etc)

We believe that in addition to the cost issue and material change to reporting processes mentioned above this change represents a fundamental change to how reporting is carried out and as such an initial framework consultation done prior to the detailed CP we respond to would have been beneficial for scoping out feasibility and for simplifying implementation on short timescales.

For further information on this submission please consult the main document attached (BBA01-#466583-v12-Pillar_3_5th_September.docx) and/or contact Nemanja Eckert, Policy Director, BBA.
Nemanja Eckert
B