DB pension redress is the FCA's prescribed mechanism for compensating consumers who were advised to transfer out of a defined-benefit pension scheme on unsuitable advice. The calculation under FCA DISP App 4 is not a generic loss-assessment exercise: it is a structured methodology with prescribed assumptions, prescribed tables, and a procedural sequence that has to be followed for the answer to be defensible. The note below sets out what the calculation does, what it requires, where it gets contested, and what makes it defensible — the methodological position behind the firm's Redress Solution.
The starting point is the regulatory anchor. FCA DISP App 4 is the rulebook section that prescribes how redress is to be calculated for unsuitable defined-benefit pension transfer advice. The current form of DISP App 4 replaced earlier guidance (FG17/9), and is itself subject to periodic refresh by the FCA: the prescribed actuarial assumptions, the discount-rate methodology, and the treatment of tax adjustments are all set out by the regulator and updated as economic conditions and FCA policy evolve. For the firm's purposes, the operative rule is whichever version of DISP App 4 was in force at the date the calculation is performed.
What the calculation produces, in simple terms, is a number: the redress amount due to the consumer. What sits behind that number is a structured methodology that takes scheme records, member data, and the facts of the original advice, and applies the prescribed assumptions to produce a defensible answer. The number is not the methodology, and the methodology is not the number; both are required for the calculation to hold up under the kinds of scrutiny redress calculations attract — skilled-person review, regulator enquiry, Ombudsman referral, or court proceedings where the redress amount is itself contested.
What FCA DISP App 4 requires.
DISP App 4 sets the framework: a consumer who received unsuitable advice to transfer from a defined-benefit scheme to a personal pension is entitled to redress that puts them, so far as is reasonably possible, in the financial position they would have been in had the unsuitable advice not been given. The redress is the difference between two situations: the counterfactual (the consumer remained in the DB scheme) and the actual (the consumer is in the personal pension that the unsuitable advice led to).
That sounds simple enough as a statement of principle. The complexity sits in operationalising it. The counterfactual has to be quantified: what would the DB scheme have paid, when, with what escalation, taxed at what rate, in what circumstances? The actual situation has to be valued: what is the personal pension fund worth at the valuation date, what investment return has it generated, what tax has been or will be paid on its drawdown? And the two have to be reconciled at a single point in time using prescribed assumptions about future investment returns, prescribed mortality tables, prescribed inflation projections, and prescribed economic-environment data drawn principally from the Bank of England.
The prescription matters. DISP App 4 is not a methodology the calculator chooses; it is the methodology the FCA has chosen. Where the calculator has discretion, it is usually narrow: choices about marginal tax rates have to be agreed between respondent and consumer; choices about means-tested benefits adjustments have to be investigated and applied; choices about complex scheme features (commutation, lump-sum options, escalation rates) have to be made consistently with the scheme's actual rules. But the core methodology — the discount rate, the prescribed assumptions, the tabular basis — is set by the regulator.
The calculation chain, from input to output.
A DB pension redress calculation has a standard shape, regardless of the case complexity. The inputs are the scheme records (benefit specification, member data, accrual history, scheme rules including escalation and commutation provisions) and the personal pension data (transfer value at the date of transfer, investment performance subsequent to transfer, current fund value, any drawdown that has taken place). Documentation of the original advice is also relevant where the redress calculation depends on agreed facts about what the advice was.
The counterfactual valuation produces the value of the DB scheme benefits at the valuation date used for the calculation — conventionally the first day of the quarter in which the calculation is performed. This requires projecting the benefits the consumer would have received had they remained in the scheme, valuing those benefits on the DISP App 4 prescribed basis, and adjusting for the fact that the consumer may have already passed the scheme's retirement age (an actual-loss case) or may not yet have reached it (a prospective-loss case).
The actual situation is valued by taking the personal pension fund as it stands at the valuation date. For prospective-loss cases there is no past-payment complication; for actual-loss cases, the calculation has to factor in historical payments under both the counterfactual and actual scenarios, with interest roll-up and income-tax adjustments where applicable.
The Primary Compensation Sum is the difference between the two: counterfactual value less actual value, with the tax and other adjustments DISP App 4 prescribes. Where redress is paid in cash rather than as augmentation of the personal pension, a tax adjustment reduces the cash amount to reflect the income-tax position the consumer would have faced on the equivalent DB scheme payments. Where the time between the valuation date and the date of actual payment is non-trivial, an Additional Compensation Sum is calculated to bring the redress forward. Where complications around historical payments or means-tested benefits arise, a Secondary Compensation Sum captures the adjustment.
The output is, in the end, the redress amount the consumer is due — but presented in the form DISP App 4 expects: a Primary Compensation Sum (the core), with Secondary and Additional Compensation Sums as separate identified components, each with its own working visible.
Where the calculation gets contested.
The contestable points in a DB pension redress calculation are well-known to practitioners. The marginal income-tax rate to be applied to historical and prospective payments is an agreed item between respondent and consumer — where the consumer's tax circumstances are unusual, this is where parties most often disagree. The treatment of means-tested state benefits is another — whether the consumer's entitlement to such benefits is affected by the redress, and whether the redress amount needs to be reduced or restructured to avoid adverse benefit interaction.
Discount rate questions can also be contested at the margin, particularly in unusual market conditions. The prescribed methodology under DISP App 4 uses Bank of England gilt yields as the principal economic-environment input; in periods of yield volatility, the choice of measurement date can produce calculations that move materially over a short window. The procedural answer (use the date prescribed by the methodology) is usually clear, but the substantive question (whether the prescribed date produces a result that the regulator and the Ombudsman would consider fair) is one practitioners need to think through.
Treatment of complex scheme features is the third contestable area. Where the DB scheme had non-standard escalation, commutation, or lump-sum provisions, the calculation has to apply those provisions consistently with the scheme's actual rules — not with assumed standard features. Pension Increase Exchange options, RPI-versus-CPI escalation indices, partial-cash-commutation pathways, and bridge pensions for early retirement are examples where the calculation can produce materially different answers depending on the assumptions about scheme rules. The defensible answer is the one rooted in the scheme's specific rules, with the working visible.
Single-case work versus past-business reviews.
The same methodology applies to a single-case redress calculation and to a high-volume past-business review. What differs is the scale, the data architecture, and the audit-traceability standard. A single-case calculation is one consumer, one set of scheme records, one personal pension fund — the calculation chain is the same as for a population, but the operational complexity is contained.
A past-business review is different. The FCA-regulated firm is required to assess the suitability of its advice across a defined customer population — sometimes following a skilled-person review under FCA section 166, sometimes on a voluntary basis as part of supervisory engagement, sometimes pre-emptively to manage the firm's regulatory exposure. The redress calculation operates at population scale: hundreds or thousands of cases, each with its own scheme records and personal pension data, calculated to the same methodology, audited at population level.
The audit-traceability requirement is different at scale. For a single case, the calculation chain has to be reconstructible end-to-end — what inputs were used, what assumptions were made, what intermediate calculations produced what numbers, how the final figure was derived. For a population, the same is required for every case, plus a layer above: sample-reproducibility evidence, methodology-consistency demonstration, and the kind of documentation a skilled person, an FCA supervisor, or an internal audit function can rely on to form a view that the population-level calculation was done properly.
This is where the firm's platform-led approach matters. A single-case DB pension redress calculation can be done on a spreadsheet by a competent actuary; a population-level redress calculation generally cannot — not because the methodology is different, but because the audit-traceability standard a population engagement requires is hard to reach from a spreadsheet-of-spreadsheets. The platform produces the same calculation receipt for every case, on the same calculation chain, with the same audit-trail format, regardless of population scale.
What audit-traceability means in this context.
Audit-traceable, as the firm uses the term, means that any qualified third party — FCA supervisor, skilled person, internal audit, Ombudsman referee, opposing expert in court — can reconstruct the calculation from the published record. Every input is recorded. Every assumption is recorded. Every intermediate calculation step is recorded. Every output is recorded with its provenance.
The Check My Calculation™ verification function operationalises this. Given a calculation identifier, the function returns the reproducibility record: the inputs the calculation was based on, the methodology version that was applied, the assumptions used, and the calculation chain that produced the output. A third party reviewing a CMC™ artefact can verify the calculation without re-doing it from scratch — the working is the audit, not just supporting evidence for an audit done separately.
In the context of DB pension redress, audit-traceability matters in three distinct ways. First, at single-case level, where a consumer or their adviser challenges the redress amount, the audit-traceable record is the basis on which the challenge can be examined and either resolved or escalated. Second, at past-business-review level, where a skilled person or FCA supervisor is reviewing the population engagement, the audit-traceable record is the basis on which the population-level work can be assessed for consistency and methodology integrity. Third, at regulator-engagement level, where the calculation has produced a result that the regulator is querying, the audit-traceable record is the basis on which the firm can defend the calculation directly — not through reconstruction, but by reference to the published record.
What the calculation cannot do.
The calculation under FCA DISP App 4 has bounds, and it is as important to be clear about what the calculation does not do as about what it does. The calculation cannot determine whether the original advice was unsuitable. Suitability assessment is a separate exercise, conducted under different FCA rules (principally COBS), and produces a different output (a suitability finding) before the redress calculation is engaged at all. The redress calculation takes unsuitability as given; it does not establish it.
The calculation cannot factor in subsequent investment performance in any way other than the prescribed methodology allows. If the consumer's personal pension has performed extremely well or extremely badly since the transfer, this fact is reflected in the actual-situation valuation through the current fund value — but the prescribed methodology does not allow the calculation to be reopened on the basis that the investment outcome should change the redress amount. The calculation is anchored at the date of transfer and the valuation date, not at the date of the consumer's eventual retirement.
The calculation cannot resolve disputes about facts. Where the parties disagree about what the original advice said, or what the scheme rules actually provided, or what the consumer's circumstances were at the date of advice, the calculation cannot resolve these. Disputed facts have to be agreed between the parties (or determined by the Ombudsman or court) before the redress calculation can produce a single defensible number. The calculation runs on agreed facts; where the facts are disputed, the calculation may produce alternative scenarios, but it does not arbitrate.
The calculation cannot compensate for non-financial loss. The redress regime under DISP App 4 is concerned with the financial position the consumer would have been in but for the unsuitable advice. Inconvenience, distress, and other non-financial harms are outside the redress calculation; they are handled, where they arise, through separate FCA mechanisms (compensation for distress and inconvenience may be available alongside redress, but is determined separately).
What this means in practice is that a DB pension redress calculation is not a number a calculator produces. It is the output of a methodology applied to facts, with prescribed assumptions, audit-traceable workings, and a procedural sequence that has to be respected for the answer to be defensible. When that methodology is right, the calculation defends itself — in front of the consumer, in front of the regulator, in front of the skilled person, in front of the court. When it isn't, the defects show up under cross-examination, in skilled-person review, or in regulator enquiry — and the cost of correcting at that point is materially higher than the cost of getting it right the first time.
The firm's position on DB pension redress is to give the calculation that defensibility from the outset — through the methodology, through the platform that produces an audit-traceable record for every case, and through the senior actuarial judgement that is responsible for the calculation end-to-end. The Redress Solution is the productised offering through which the firm delivers this; the methodology is the same whether the engagement is one case, a hundred, or several thousand.
For the firm’s productised FCA-regulated redress offering — single-case calculations, past-business reviews, and population-scale remediation — see Redress & remediation.
Last updated May 2026.