Ordinary stress testing asks: if rates move by 200 basis points, what happens to EVE? That is useful, but it answers a question the regulator set, not the one the board actually worries about. Reverse stress testing flips it around and asks the more useful question:
“How far can rates move before we breach our EVE limit?”
That single number — the breakeven shock — tells the ALCO how much headroom the balance sheet has today, in the units they care about.
The regulatory outlier test flags a bank when the change in EVE under a prescribed shock exceeds 15% of Tier 1 capital. Reverse stress solves that condition backwards: instead of fixing the shock and reading the ΔEVE, it fixes the ΔEVE at the limit and solves for the shock.
Because ΔEVE moves monotonically with the size of a parallel shock, the solution is found efficiently by binary search:
Parallel-up and parallel-down are solved separately, because a balance sheet rarely has symmetric headroom — the shock that hurts EVE on the way up is seldom the same size as the one that hurts on the way down.
The result sits inside the same stress capsule that produces EVE and NII, so it is computed on exactly the same cashflows and curves — not a separate, disconnected calculation.
Open the interactive reverse-stress tool → · See the full EVE / NII / reverse-stress capsule →
Illustrative methodology on synthetic data. No client data, production code or parameters are reproduced.