Start from what the PRA will measure. SS24/15 asks for a range of severe but plausible scenarios and for the lowest point of cumulative stressed net cashflows, both within the 30-day LCR horizon and across the survival days. As we read it, the statement sets no single fixed survival period: the firm’s risk appetite defines the stress it aims to survive, and the Pillar 2 baseline is survival throughout the 30-day granular LCR scenario on a consolidated currency basis.
SoP1/18 adds the PRA’s own lens. It tests cash flow mismatch risk, insufficient liquidity from HQLA and other inflows to cover outflows daily, through the 30-day granular scenario and benchmark retail-only and wholesale-only scenarios over 90 days. The low point sits inside the window: a peak need before day 30 that is larger than the day-30 figure. A model that reports only day 30 cannot see it, and that is the gap we find most often.
A stress test set that will hold up covers:
- the 30-day granular scenario with a daily cumulative profile, then the 90-day benchmarks, then the firm’s own scenarios, including the one its risk appetite says it survives
- behavioural assumptions for deposits set in one place, with the evidence behind each
- contingency funding actions sized and timed against the gap each scenario leaves, matched to the recovery plan’s options rather than duplicating them
Then test the data. SS24/15 expects a firm to be able to produce key data on buffers, cashflows, wholesale counterparties and FSCS balances during a stress. We ask the treasury team to produce that pack once, on a normal day, before the ILAAP says they can.
Sources: PRA SS24/15, chapter 2 (paras 2.10 to 2.18), paras 3.12 and 6.3; PRA SoP1/18, paras 3.2, 3.5, 3.8 and 3.13.