PRA Pillar 2: Internal Stress Test Reporting
On 30th April 2018, the PRA published its Policy Statement (PS) PS8/181 on Pillar 2 reporting, finalising its Consultation Paper (CP) CP25/17.2 The PS requires firms to submit a new return (PRA111) to capture internal stress testing data used for Pillar 2B assessments, currently included in firms’ Internal Capital Adequacy Assessment Process (ICAAP) documents.
The PS applies to banks, building societies and PRA-designated investment firms, with total assets >= £5 billion, at the ICAAP consolidation level and frequency of submission.
Importantly, the PRA may ask firms that are not covered by the definition above to provide information in this format on a case by case basis. These requirements take effect from 1 October 2018.
The PRA111 return provides a standardised reporting framework for firms’ internal Pillar 2B stress test data. Until now, this information was only provided in the firms’ ICAAP documents at varying levels of granularity and transparency.
The direction of travel is unmistakable: We see this as part of a broader theme of regulatory focus on structured information to be provided by firms for the PRA to be able to make consistent and informed judgements about the point in time and forward looking risk profiles of firms. This thought process led to the introduction of standardised reporting for key Pillar 2A risks (FSA071-082 reports) and now PRA111 for Pillar 2B reporting.
This development is also consistent with the standardised reporting templates provided by the Bank of England (BOE) for its annual macro-economic stress testing exercise for major banks in the UK. The consistency of these templates with the PRA111 may be a relief for the larger banks; however, the low threshold and the discretion available to the PRA means that mid-size and smaller firms will have to take potential significant steps to comply with the requirements.
Finally, this PS seems to resonate with the latest EBA GL on Stress Testing which states that supervisors shall benchmark the results/assumptions of internal stress tests to supervisory stress tests.3
Data needs to be provided for the stress scenario used by the firm to size the PRA buffer. The time horizon will be three to five years, in line with firms’ ICAAP projections.
The return outlines the following key data categories:
Data categories 2-7 are to be provided for each year of the planning horizon under base and stress assumptions; for categories 3-7, data needs to be provided on a pre- and post-management action basis.
Within remuneration reporting in the P&L impacts, the inclusion of the fixed/variable split is interesting; this aspect would likely inform the governance element of SREP. PRA would have visibility over the compliance with variable remuneration capping requirements over time. Furthermore, a view of the business plan and projected variable remuneration may provide PRA with some interesting insights on longer term risk appetite and risk-taking.
It is also interesting that the paper does not require any split between the various IFRS9 stages for reporting impairment information.
We see a number of key implications for firms:
- Strategic and capital planning: There are likely to be some differences between the level of granularity that Firms currently conduct their base case strategic and capital planning in comparison with what is expected by the PRA111 return, particularly firms which are not a part of the annual BOE stress testing exercise. Firms will have to undertake further work to address this potential challenge.
- Ownership: We expect further discussions between risk and finance functions regarding ownership of the return or parts of the return based on practices adopted by the firms.
- Internal systems and controls: Firms are expected to have robust internal processes and controls to ensure that the data used for internal stress testing is accurate, comprehensive and complete. Firms will need to consider additional controls and governance steps for the accurate population of these reports. These discussions need to happen in an accelerated manner given the submission timelines.
- Data Reconciliation and Management: The return expects data to be provided across exposure class splits that may be different from how firms currently manage their portfolios and data.As a results, Firms may find it challenging to provide all data fields at the expected level of granularity from Day 1 and will have to consider, for the first few iterations, any need to develop short term/manual solutions to populate the return with a clear path to a systematic solution where this is possible. Firms will also need to ensure that data provided is consistent with other internal and external reports and that any internal/external audit or risk and control assessments (RCA) processes are appropriately considered.
- Tactical and Strategic Solutions: In the medium to long term, firms will need to consider if strategic and capital planning processes need to be enhanced to develop forecasts at the required level of granularity.
- ICAAP Submissions: Firms will need to consider the impact of populating PRA111 on the overall ICAAP development, stress testing and governance processes. Compared to reporting for P2A risks, reporting for P2B may impose potentially significant burdens on the teams conducting the stress test exercises. Teams will need to capture the relevant data points from a number of sources including strategic plans, current stress testing engines/models/excel sheets and capital returns to populate this report. At the same time, there will be an expectation that the outputs are tabled alongside the ICAAP as part of the governance, review and challenge process.
- Management Actions: There are a range of practices adopted by firms in relation to the consideration of management actions in the quantification of P2B estimates, particularly around the distinction between mitigating actions that a firm may take on a BAU basis during the stress period (e.g. dividend reduction) and those that are taken as a “post-fact” adjustment to the stressed outcomes (e.g. reduction of discretionary costs). Firms may need to reconsider how they integrate management actions in their quantification processes to report appropriately.
In addition, we see some other considerations for firms and the PRA:
Any peer analysis conducted by the PRA would need to consider the differences in the definition and calibration of internal scenarios, varied timings and quantum of management actions, etc. to provide a meaningful basis for comparison.
We expect the supervisory trend towards standardisation of key metrics and reports to continue in the future.
To this end, the PRA111 return provides a standardised format for the PRA to assess firms’ internal stress tests. It also provides firms’ internal management with a consistent view, comparable over years, of the firms’ internal assessment of P2B capital requirements.
Finally, the PRA expects that implementation of PRA111 would reduce the number of additional requests and clarifications sought by the PRA during their assessment of firms’ P2B stress tests.
3https://www.eba.europa.eu/documents/10180/2006890/Consultation+Paper Para 124 states: “…When challenging scenarios, assumptions, and outcomes of institutions’ stress tests done for ICAAP and ILAAP purposes, competent authorities should use, where appropriate, the outcomes, scenarios and assumptions used in the supervisory stress tests, including relevant regional stress test exercises done by various authorities…”
For more details, please contact the authors of this blog.