The recovery plan readiness checklist: ten areas a plan is tested on

Recovery plans are rarely found to be wrong in one decisive way. They decay: options expire as the group changes, thresholds drift as the balance sheet moves, named individuals leave, and the document is refreshed each year by updating the figures rather than re-examining the reasoning.

The ten areas below are where a plan is most consistently tested — by the board, by a rehearsal conducted against the clock, and by a supervisor. Each is stated as what a sound plan can show, followed by the symptom that gives a weak one away.

The pattern across all ten is that the defects are found by asking for specifics, not by reading the document.

Strategic analysis and critical function mapping#

A sound plan can show which business lines are core, which functions would matter beyond the institution’s own walls if they stopped, how licences and capital sit across legal entities, and which systems, staff and contracts are shared between them.

The symptom of a weak answer is a strategic analysis that reads as a corporate profile, or an inability to say whether a business earmarked for disposal is actually separable — which turns the option menu into a list of intentions.

Indicators and calibration#

A sound plan can show, for each threshold, how it was derived — the option it is meant to protect, that option’s execution lead time, and the deterioration rate it was set against.

The symptom of a weak answer is thresholds sitting a short distance above regulatory minimums, which fire when the actions they command already require markets the conditions have closed; or the opposite, thresholds breached routinely with a standing explanatory comment attached, which is how a framework loses its authority before it is ever used.

Escalation and the governance playbook#

A sound plan can show who is authorised to convene the recovery forum and how quickly, what that forum may decide without further approval, what happens when a named individual is unreachable, and how decisions are recorded when the usual secretariat is not operating.

The symptom of a weak answer is escalation described in the language of ordinary committee process — quarterly cycles, standard notice periods, full board approval for an option with a two-day execution window.

Delegations that cannot be exercised at speed are not delegations.

Option quantification#

A sound plan can show, for every option, the capital or liquidity it generates, the calendar time from decision to effect including approvals, what it depends on, and what it costs — including the permanent damage to future earnings where the option sells part of the franchise.

The symptom of a weak answer is an option menu of descriptions rather than quantities, or execution times measured in the time the transaction itself takes rather than the time the institution would actually need to reach the point of transacting.

Option diversity and independence#

A sound plan can show that its options do not all depend on the same condition holding — that they span capital generation, liquidity generation, risk-weighted asset reduction, resource conservation and structural action, and that they are ordered from reversible and cheap to permanently transformative.

The symptom of a weak answer is three capital options that each require a market willing to buy bank assets at a reasonable price, which is one option written three times, and which the scenario that would trigger them has already removed.

Scenario testing#

A sound plan can show its options tested against scenarios severe enough to threaten viability, including both a slow capital erosion and a fast liquidity path, and can produce the resulting gap analysis — which scenarios the current menu cannot answer and what would have to be arranged in advance.

The symptom of a weak answer is a plan tested only against scenarios the institution comfortably survives, or one whose scenarios bear no relationship to the reverse stress testing the institution has already performed, which suggests the two exercises have never been in the same room.

The liquidity leg and contingency funding#

A sound plan can show a survival horizon under its own stress assumptions, funding sources ranked by how quickly they would become unavailable, collateral identified and confirmed as eligible and unencumbered, and arrangements put in place before the conditions that would make them impossible to arrange.

The symptom of a weak answer is a contingency funding plan listing facilities that have never been tested, collateral whose eligibility has not been confirmed, or a survival horizon computed on assumptions about depositor behaviour that nobody has revisited since the framework was written.

Communication#

A sound plan can show the audiences it would address, who speaks to each, in what sequence, and what has been drafted in advance.

The symptom of a weak answer is a communication section consisting of the principle that communication is important, or a sequence in which a supervisor would learn of a recovery action from the market — which conveys something about the institution well beyond the action itself.

Rehearsal#

A sound plan can show a simulation in which the people who would actually take the decisions worked through a scenario against the clock, and can produce the findings — because a rehearsal that produced none was not a rehearsal.

The symptom of a weak answer is a plan approved annually and never exercised, or an exercise conducted as a presentation to the participants rather than a decision they had to take.

This is the single most diagnostic area on the list: institutions that rehearse discover that half the option menu is theoretical, and institutions that do not, discover it later under worse conditions.

Maintenance and ownership#

A sound plan can show when each section was last re-examined rather than merely re-dated, who owns each part, and what structural changes since the last review have been reflected — a disposal, an acquisition, a funding line withdrawn, a licence moved.

The symptom of a weak answer is a plan whose option menu still names a subsidiary sold two years ago, whose escalation lists individuals who have left, or whose annual refresh updated the figures without anyone asking whether the reasoning still held.

The test that does most of the work#

Across the ten, one test does more work than the others: give the plan to the people who would have to use it, put them under a clock, and see what breaks. Everything on this list that is wrong will surface within a few hours, and it will surface while it is still cheap to fix.

A plan that has survived the test that does most of the work and been amended afterwards is a capability. A plan that has only ever been approved is a submission.