The Business Indicator is the measure of size that sets a UK bank’s Pillar 1 operational risk capital from 1 January 2027. It is the sum of three components taken mainly from the income statement, each averaged over three years: interest, leases and dividends; services; and the financial component. The bank multiplies it by marginal coefficients of 12%, 15% and 18% to give the Business Indicator Component, and because the PRA has set the Internal Loss Multiplier to one, that component is the own funds requirement (Operational Risk Part rules 4.1, 5.2, 5.7 and 5.9). A bank with a Business Indicator of £2 billion needs £273.6 million of own funds for operational risk.
What goes into the Business Indicator?
Three components, each built from line items listed in Annex 1 of the Operational Risk Part and each a three-year average (rules 5.2 to 5.4).
| Component | Formula, using three-year averages | What it captures |
|---|---|---|
| Interest, leases and dividend component | The lower of absolute net interest income and 2.25% of interest earning assets, plus dividend income | Lending and deposit taking, with net interest income capped for high-margin books |
| Services component | The higher of other operating income and other operating expense, plus the higher of fee income and fee expense | Fee, commission and other operating business |
| Financial component | Absolute net profit or loss on the trading book, plus absolute net profit or loss on the banking book | Trading and other financial results |
Source: Operational Risk Part rule 5.3, with the items defined in Annex 1 Tables A to C. Annex 1 Table D lists what stays out, including administrative expenses such as staff costs and outsourcing fees for non-financial services, depreciation and corporate tax.
The services and financial components use the larger figure or the absolute value, so a loss-making fee business or a trading loss still adds to the indicator. The PRA considered capping the services component and decided against it, because a cap “would result in an underestimation of the scale of a firm’s operations” (PS17/23 paragraph 5.12).
How is the three-year average worked out?
Rule 5.4 sets four points that matter in practice.
- Absolute values come first. Net items are made absolute year by year, and only then averaged (rule 5.4(2)). Trading results of a £10 million gain, a £10 million loss and a £10 million gain average to £10 million, not £3.3 million.
- Audited figures where they exist, business estimates where they do not (rule 5.4(3)).
- At the financial year end the current year must be included in the average (rule 5.4(4)). PS1/26 added this limb, and said that where audited figures for the current year are not available, firms should use business estimates rather than outdated audited data (paragraph 2.13).
- A firm in operation for less than three years may use forward-looking estimates (rule 5.4(1) and (2)).
Acquisitions, mergers and disposals during the three years are included by default (rule 5.5(1)). A firm can apply for permission to exclude divested activities where the three-year average would give a biased estimate (rule 5.5(2)), a change the PRA made in its near-final rules (PS17/23 paragraph 5.4). A firm that has sold a business line recently should look at this before its first 2027 calculation.
How does the Business Indicator become a capital requirement?
The indicator is multiplied by marginal coefficients that rise with size (rules 5.7 and 5.8).
From income statement to operational risk capital in six steps
- Build the three componentsInterest, leases and dividends; services; and the financial component, from the Annex 1 line items (rules 5.2 to 5.4).
- Take three-year averagesMake net items absolute year by year and only then average them, including the current year at year end (rule 5.4).
- Sum to the Business IndicatorAdd the three averaged components together to give the Business Indicator.
- Apply marginal coefficients12% up to £0.88 billion, 15% above £0.88 billion up to £26 billion, 18% above £26 billion (rules 5.7 and 5.8).
- Read off the own funds requirementWith the Internal Loss Multiplier at one, the result is the requirement: 12% × £880m + 15% × £1,120m gives £273.6m (rule 5.9).
- Multiply by 12.5This gives the risk-weighted exposure amount, £3,420m for the £2 billion example (Article 92(4)(b)).
| Bucket | Business Indicator | Marginal coefficient |
|---|---|---|
| 1 | Up to £0.88 billion | 12% |
| 2 | Above £0.88 billion, up to £26 billion | 15% |
| 3 | Above £26 billion | 18% |
The coefficients work like income tax bands: each rate applies only to the slice of the indicator inside its band. The sterling thresholds are the Basel Committee’s €1 billion and €30 billion, converted at £1 = €1.14 and rounded (PS17/23 paragraph 7.2). The result is an own funds requirement. Multiplying it by 12.5 gives the risk-weighted exposure amount for operational risk (Required Level of Own Funds (CRR) Part, Article 92(4)(b)).
Two worked examples, with the Internal Loss Multiplier at one:
| Business Indicator | Calculation | Own funds requirement | Risk-weighted exposure |
|---|---|---|---|
| £500 million | 12% × £500m | £60.0m | £750m |
| £2 billion | 12% × £880m + 15% × £1,120m | £273.6m | £3,420m |
The second bank pays an effective rate of 13.7% of its indicator. The PRA’s own consultation example runs the same arithmetic for a Business Indicator of £35 billion and reaches £5.49 billion (CP16/22 paragraph 8.18, footnote 4).
Why can a group’s requirement exceed the sum of its subsidiaries?
Because the rate rises with size. At consolidated level the Business Indicator is calculated on fully consolidated figures, netting intragroup income and expenses, while each subsidiary uses its own figures (rule 5.6). A group whose combined indicator crosses £0.88 billion pays 15% on the excess, even if no subsidiary does on its own. The PRA says this is “an intended outcome of the BIC varying based on the size of the firm” (PS17/23 paragraph 5.13).
Why do the bank’s own losses not count?
The Basel standard scales the Business Indicator Component by an Internal Loss Multiplier driven by a loss component of 15 times a bank’s average annual operational losses over the previous ten years, and lets national supervisors set the multiplier to one for all banks (Basel III: Finalising post-crisis reforms, operational risk paragraphs 9 and 12). The PRA did so (rule 5.9). It argued that losses are fat-tailed and that “past events (particularly over a lengthy historical period) are generally not good predictors of future losses” (PS17/23 paragraph 5.20), and that the size of a firm is “the dominant differentiator of operational risk” (CP16/22 paragraph 8.28).
A clean loss record therefore earns no reduction in Pillar 1, and a poor one adds nothing there. Losses still count in Pillar 2A, where the PRA weighs the firm’s historical losses and the design and severity of its scenario analysis (SoP5/15 paragraph 4.9A). Our article on how the PRA sets Pillar 2A covers that side.
Does a bank still need to collect loss data?
Yes. Chapter 7 of the Operational Risk Part sets loss data requirements for the firms the Part applies to, and we found no size threshold in it. The main rules:
- A ten-year observation period, or exceptionally no less than five years (rule 7.1(2)).
- A minimum threshold of £20,000 for including a loss event (rule 7.1(4)).
- Gross losses, non-insurance recoveries and insurance recoveries identified separately, with recoveries used to reduce losses only after payment is received (rule 7.2(1) and (2)).
- Losses mapped to the seven Level 1 event types in Annex 2, from internal fraud to execution, delivery and process management (rule 7.1(3)).
- Credit-related losses already reflected in credit risk-weighted exposures kept out of the data set, and operational losses related to market risk treated as operational risk (rule 7.1(7) and (8)).
Disclosure brings its own threshold. An institution with a Business Indicator of £880 million or more discloses its annual loss data for each of the previous ten years, and institutions other than SDDTs disclose each of the Business Indicator sub-items (Disclosure (CRR) Part, Articles 433a to 433c and 446). Insurance does not reduce the requirement in either pillar (PS17/23 paragraph 5.12; PS15/26 paragraph 3.9).
What should a bank check before its first calculation?
- Map the general ledger to the Annex 1 items for each of the last three years, and reconcile the result to the audited accounts.
- Test the Table D exclusions line by line. Staff costs and outsourcing fees sit in different places in different ledgers.
- Run the year-end calculation with the current year included, so the 2027 capital plan reflects it.
- Decide whether a recent disposal justifies applying for permission to exclude divested activities.
- Bring loss data collection up to the Chapter 7 standard, including recoveries and the reference dates in rule 7.1(5).
- Rewrite the ICAAP’s operational risk section for a Pillar 1 figure now driven by the Business Indicator (and, for a former AMA firm, no longer by its own losses), and agree with the supervisor how Pillar 2A will be rebased.
Our operational risk capital work sizes the new Pillar 1 charge against today’s and rebuilds the Pillar 2A scenario analysis. For the rest of the package, see what Basel 3.1 changes for UK banks and updating your ICAAP for Basel 3.1. If the work needs a permanent owner, our head of operational risk search finds one. Terms are defined in our glossary.
Questions readers ask
When does the Business Indicator approach apply?
From 1 January 2027, with the rest of the UK Basel 3.1 package (PS1/26 paragraph 1.23). It replaces the basic indicator, standardised and advanced measurement approaches (PS17/23 paragraph 5.2; CP16/22 paragraph 8.5), and the PRA withdraws SS14/13 on operational risk the same day (PS1/26 paragraph 2.14).
Does the Business Indicator apply to SDDTs?
Yes. The SDDT capital regime builds its Pillar 1 on the Basel 3.1 standardised approaches to credit and operational risk (PS20/25 paragraph 1.8), and the final policy in PS4/26 made no substantive change to that (paragraph 1.12). Our SDDT page covers who qualifies.
Can a bank lower its operational risk capital with a good loss record?
Not in Pillar 1. With the Internal Loss Multiplier at one, the requirement depends only on the Business Indicator (rules 4.1 and 5.9). A strong loss record and good scenario analysis can still matter to the PRA’s Pillar 2A judgement (SoP5/15 paragraphs 4.9A and 4.12).
Will total operational risk capital go up?
The PRA’s intention is that total Pillar 1 plus Pillar 2A operational risk requirements stay unchanged for most firms, by adjusting Pillar 2A for the change in Pillar 1 (PS17/23 paragraph 6.21 and footnote 32). The adjustment is limited to the firm’s existing Pillar 2A operational risk requirement, so a firm with none gets no offset for a Pillar 1 increase. In May 2026 the PRA said its off-cycle review of firm-specific requirements was under way (PS15/26 paragraph 3.26).
Sources: PRA Rulebook, Operational Risk Part (version effective 1 January 2027), rules 4.1, 5.2 to 5.9, 7.1 and 7.2 and Annexes 1 and 2; PRA Rulebook, Required Level of Own Funds (CRR) Part, Article 92(3)(e) and 92(4)(b); PRA Rulebook, Disclosure (CRR) Part, Articles 433a to 433c and 446; PRA PS17/23 (12 December 2023), paragraphs 5.2, 5.4, 5.12, 5.13, 5.20, 6.21 (and footnote 32) and 7.2; PRA CP16/22, chapter 8 (30 November 2022), paragraphs 8.5, 8.18 and 8.28; PRA PS1/26 (20 January 2026), paragraphs 1.23, 2.13 and 2.14; PRA PS15/26 (28 May 2026), paragraphs 3.9 and 3.26; PRA SoP5/15 (version effective 1 January 2027), paragraphs 4.9A and 4.12; PRA PS20/25 (28 October 2025), paragraph 1.8; PRA PS4/26 (20 January 2026), paragraph 1.12; Basel Committee, Basel III: Finalising post-crisis reforms (December 2017), operational risk paragraphs 9 and 12. Accessed 1 October 2026.
How we research and check our articles: our editorial policy.