The Rulebook requirement is short. A recovery plan must include “a framework of indicators” that identifies the points at which recovery actions may be taken, and the indicators must be “capable of being monitored easily” (Recovery Plans Part rules 6.2 and 6.3). SS9/17 supplies the detail, and since 1 April 2026 an SDDT reviews its plan at least every two years instead of every year (rule 4.2).
Which indicators does SS9/17 expect?
A range wider than the regulatory ratios, with early warning indicators set ahead of the recovery triggers (SS9/17 paragraph 2.47). SS9/17 also expects firms to follow the EBA guidelines on the minimum list of recovery plan indicators, which make capital, liquidity, profitability and asset quality mandatory and add market-based and macroeconomic indicators unless the firm justifies leaving them out (paragraph 2.50; EBA/GL/2015/02 paragraphs 11 and 12).
| Category | Minimum indicators (EBA Annex II) | Status |
|---|---|---|
| Capital | CET1 ratio; total capital ratio; leverage ratio | Mandatory |
| Liquidity | Liquidity coverage ratio; net stable funding ratio; cost of wholesale funding | Mandatory |
| Profitability | Return on assets or return on equity; significant operational losses | Mandatory |
| Asset quality | Growth rate of gross non-performing loans; coverage ratio (provisions to non-performing loans) | Mandatory |
| Market-based | Rating under negative review or downgrade; CDS spread; stock price variation | Included unless shown not to be relevant |
| Macroeconomic | GDP variations; sovereign CDS | Included unless shown not to be relevant |
Source: EBA Guidelines on the minimum list of qualitative and quantitative recovery plan indicators (EBA/GL/2015/02, 23 July 2015), paragraphs 11 and 12 and Annex II, as published by the Bank of England.
Each listed indicator should be included or its absence explained, and firms should add indicators of their own (SS9/17 footnote 12). An unlisted bank has no share price and usually no traded CDS, so it has to justify the substitutes it uses; the EBA’s illustrative list, which includes cost of total funding, concentration of funding sources and net interest margin, is a starting point (EBA/GL/2015/02 Annex III).
Two further expectations shape the set. Indicators should track projections and trends as well as actual results, because the rate of change early in a stress can predict how fast it will deepen (SS9/17 paragraph 2.49). And a trigger is a prompt to consider the situation, which “should not be used as an automatic trigger for a predefined set of management actions” (paragraph 2.45). The plan should also say at what point the PRA would be told (paragraph 2.82).
How should capital indicators be calibrated against buffers?
Early enough that the slowest credible recovery option can still be completed, and justified against the relevant buffers. SS9/17 paragraph 2.51 names the inputs: the range of credible options, the time each takes including governance and regulatory approvals, the firm’s risk appetite and tolerance, and, for capital and liquidity indicators, “the relevant buffers”.
Paragraph 2.52 adds a capital point: firms should allow for automatic maximum distributable amount restrictions and “calibrate their capital indicators accordingly”. The EBA guidelines approach the same question from the other side, asking for thresholds at “a sufficient distance from a breach of the capital requirements”, measured without the buffers (EBA/GL/2015/02 paragraph 25). A calibration that satisfies both shows, for each capital indicator, the distance to the buffer and the distance to the requirement.
A calibration table with one row per indicator makes that reasoning visible. The columns, and the paragraph of SS9/17 each answers to, are below.
| Column | What goes in it | Basis |
|---|---|---|
| Metric | The indicator, its definition, its data source and how often it is measured | SS9/17 2.46, 2.47 |
| Current and projected value | The latest reading and the business plan projection, with the trend | 2.49 |
| Risk appetite limit | The board’s limit for the metric, where one exists | 2.51 |
| Early warning level | The level that prompts escalation, and why it sits there | 2.45, 2.47 |
| Recovery trigger | The level at which the board considers invoking the plan | 2.45, 2.51 |
| Buffer and requirement | The distance from the trigger to the relevant buffer and to the requirement | 2.51, 2.52; EBA paragraph 25 |
| Option protected | The slowest credible option the trigger must leave time for, with its execution time and time to realise benefits | 2.26, 2.51 |
| Scenario evidence | When the trigger fires in each scenario, and whether that left enough time | 2.58, 2.63 |
| PRA notification | The point at which the supervisor is told | 2.82 |
What changes for an SDDT moving to the Single Capital Buffer?
The buffer that capital indicators are measured against. From 1 January 2027 an SDDT holds one Single Capital Buffer, set at no lower than 3.5% of RWAs, in place of the capital conservation buffer, the countercyclical buffer and the PRA buffer, and it leaves the automatic restrictions of the maximum distributable amount framework (PS20/25 paragraphs 1.14, 4.2 and 4.4, confirmed without substantive change by PS4/26 paragraph 1.12; SoP5/25 paragraphs 12.11 and 12.46).
That removes the anchor paragraph 2.52 relies on. The March 2026 update to SS9/17 did not touch the calibration paragraphs, which still refer to buffers and MDA restrictions; for an SDDT the relevant buffer from 2027 is the SCB. In place of the MDA point, SoP5/25 sets expectations for using the buffer (paragraphs 12.31 and 12.43 to 12.45):
- the SCB is meant to be used in a stress, and using it is not a breach of capital requirements or threshold conditions;
- it should not be used in the normal course of business or in the base plan, and the base case should meet it;
- an SDDT that identifies the need to draw on it must notify the PRA as early as possible, with the circumstances, how the buffer will be used and a capital restoration plan;
- a firm that does not meet its SCB can expect enhanced supervisory scrutiny.
A workable ladder for an SDDT therefore has three reference points instead of one: the top of the SCB, where the notification and capital restoration plan must be ready; any solvent exit capital trigger, which in the PRA’s own example sits inside the buffer and above the requirement; and total capital requirement itself. In that hypothetical case the firm expected to breach its solvent exit capital trigger three months before its total capital requirement (SoP5/25, Annex, Case Study E). Early warning and recovery triggers belong above the top of the buffer, so the board weighs its options before it has to tell the PRA it is drawing down. Our article on solvent exit analysis covers that trigger.
One more point for the two-year cycle. SS9/17 still expects firms to follow the EBA guidelines, and they ask for indicators to be recalibrated “when necessary and at least annually” (EBA/GL/2015/02 paragraph 17). For a small firm with a simple business model whose key metrics have not moved, the governing body may decide at its biennial review that the previous year’s triggers remain appropriate (SS9/17 paragraph 2.84A); for anyone else, the calibration table is worth looking at in the year between reviews.
How do indicators link to the ICAAP reverse stress test?
The reverse stress test finds where the business fails, and the indicators have to fire well before that point. SS9/17 asks the firm to define and justify its point of near failure, to run scenarios severe enough to reach it where they remain plausible, and encourages the use of reverse stress testing to do so (paragraph 2.58 and footnote 13).
The ICAAP statements make the same link from the other side. Firms may use reverse stress testing as a starting point for recovery plan scenarios, and business failure means the point at which the market loses confidence, which “may be reached well before the firm’s financial resources are exhausted” (SS31/15 paragraphs 4.3 and 4.8; SS4/25 paragraphs 4.2 and 4.8). A reverse stress test that runs capital to zero gives the indicators nothing to calibrate against.
The EBA guidelines ask for capital indicators to be integrated into the ICAAP (paragraph 23), and SS9/17 expects the ICAAP, ILAAP and recovery plan to be consistent: similar scenarios should show broadly similar impacts, and the recovery plan should include every management action proposed in the other documents (paragraph 2.95). Where the ICAAP’s management actions and the recovery options differ, the board should be able to say which set it believes.
How many scenarios does a recovery plan need?
At least three for most banks and building societies, four for G-SIIs and O-SIIs, and two for firms their supervisor has told are eligible for Simplified Obligations (SS9/17 paragraphs 2.3 and 2.54). Every firm, Simplified Obligations firms included, should include a scenario that combines system-wide and idiosyncratic events (paragraph 2.55).
For indicators, the scenarios are the test. The plan should show which indicators trigger in each scenario and when (paragraph 2.58), chart capital and liquidity over time with and without the recovery options, marking where early warning and recovery indicators fire (paragraph 2.66), and refine the calibration before submission if the scenarios show the firm would not have been alerted in time (paragraph 2.63). Recovery capacity is stated in CET1, leverage ratio and LCR percentage points for each scenario (paragraph 2.39). The plan is then tested live, with at least one fire drill before each submission and at least one every three years (paragraph 2.72).
What changed in the March 2026 update?
The review cycle for SDDTs, with only minor edits elsewhere. SS9/17 was republished on 26 March 2026, effective from 1 April 2026, after PS10/26; its annex records that an SDDT now reviews its recovery plan at least every two years instead of at least annually, and that other changes refresh links and references and correct typing errors.
The Rulebook change is in rule 4.2, with the same two-year cycle for an SDDT consolidation entity’s group plan in rule 4.3. Plans must still be updated for any change in structure, business or financial position that could materially affect them, and the PRA expects that SDDTs which are new and growing banks will likely need to review and update their plans more often than the minimum (PS10/26 paragraph 1.12). SS9/17 now asks SDDTs to use the extra time “to develop higher quality recovery plans” (paragraph 2.84B).
Our recovery planning service calibrates indicators, runs the scenarios and holds the fire drill, and RisKIT model 900 holds triggers, thresholds and options in open Excel. For the capital side of the SDDT change, see our article on SS4/25 and the SDDT ICAAP.
Questions readers ask
Are recovery indicators the same as risk appetite limits?
No. SS9/17 expects indicators to be calibrated relative to risk appetite and tolerance, so that recovery options can be used before the firm breaches its tolerance (paragraph 2.51), with early warning indicators ahead of the recovery triggers (paragraph 2.47). Setting the recovery trigger at the appetite limit leaves no room between the two.
Must we use every indicator on the EBA list?
Include each one or explain why not. SS9/17 expects firms to consider all the qualitative and quantitative indicators in the EBA minimum requirements, to explain any decision to leave one out, and to add indicators beyond the minimum (footnote 12).
Should the contingency funding plan use the same indicators?
Yes. SS9/17 paragraph 2.46 says firms should monitor the same set of metrics in the contingency planning framework and the recovery planning framework, so risk is monitored consistently across the firm.
How often should indicators be recalibrated?
When necessary and at least annually under the EBA guidelines, which SS9/17 expects firms to follow (EBA/GL/2015/02 paragraph 17; SS9/17 paragraph 2.50). Calibration should also be refined before a plan is submitted if scenario testing shows an indicator would have fired too late (SS9/17 paragraph 2.63).
Sources: PRA SS9/17 Recovery planning (March 2026, effective 1 April 2026), paragraphs 2.3, 2.26, 2.39, 2.45 to 2.52, 2.54, 2.55, 2.58, 2.63, 2.66, 2.72, 2.82, 2.84A, 2.84B and 2.95, footnotes 12 and 13 and the annex of updates; PRA Rulebook, Recovery Plans Part, rules 4.2, 4.3, 6.2 and 6.3; EBA Guidelines on the minimum list of qualitative and quantitative recovery plan indicators (EBA/GL/2015/02, 23 July 2015), paragraphs 11, 12, 17, 23 and 25 and Annexes II and III; PRA PS10/26 (26 March 2026), paragraph 1.12; PRA PS20/25 (28 October 2025), paragraphs 1.14, 4.2 and 4.4; PRA PS4/26 (20 January 2026), paragraph 1.12; PRA SoP5/25 (July 2026 version), paragraphs 12.11, 12.31, 12.43 to 12.46 and Annex, Case Study E; PRA SS31/15 (July 2026 version), paragraphs 4.3 and 4.8; PRA SS4/25 (July 2026 version), paragraphs 4.2 and 4.8. Accessed 10 October 2026.
How we research and check our articles: our editorial policy.
