IFoA SP4 common mistakes
IFoA SP4 · Understand the traps and how to avoid them.
In funding and valuation questions, the candidate computes a contribution rate by dividing the deficit by a single year or by the remaining average future service, without first projecting the deficit forward with interest and without allowing for the scheme's funding objective (e.g. full solvency over a stated period). The number is presented as the answer with no amortisation schedule.
Candidates latch onto the arithmetic they can do quickly and treat the deficit as a static number. They forget that the deficit accrues interest at the valuation rate, that the amortisation period is usually specified in the question, and that contributions themselves are paid through the period, so the correct approach is a level annual payment solving a present-value equation, not a simple division.
What to do instead
Before computing, write down the four inputs explicitly: deficit at valuation date, amortisation period, valuation rate of interest, and whether contributions are assumed in advance or in arrears. Then set up the equation of value (deficit equals the annuity certain or annuity due of the level contribution) and solve. State the assumption about when contributions are paid and show the resulting schedule. Practise this with both a level amount and a level percentage-of-payroll contribution, since payroll growth changes the pattern.
When asked about investment strategy, the candidate recommends an asset allocation without linking it to the scheme's liability profile, funding level, or maturity. They produce a generic 'diversified growth portfolio with some bonds' answer that would fit any scheme and ignore the scheme's duration, cash flow needs, and sponsor covenant.
Asset-liability management is the integrating skill of the exam, but candidates revise asset classes and liabilities separately and never practise joining them. Under time pressure they fall back on textbook asset-class descriptions rather than deriving the strategy from the specific liability cash flows, funding ratio, and covenant strength given in the scenario.
What to do instead
Adopt a fixed structure for every investment answer: (1) state the scheme's objectives in terms of return, risk, liquidity, and funding-level stability; (2) characterise the liabilities (duration, inflation linkage, maturity, cash-flow profile); (3) assess the sponsor covenant and contribution flexibility; (4) only then select assets, justifying each by which objective it serves and which risk it hedges. Explicitly quantify the interest-rate and inflation hedge ratios you are targeting. If the question gives a funding level and duration, use them in the recommendation.
In accounting questions the candidate conflates the IAS 19 / FRS 102 defined benefit expense components, mixing the service cost, net interest on the net defined benefit liability, and remeasurements, and puts actuarial gains and losses through profit or loss rather than other comprehensive income.
The accounting standards are revised periodically and candidates learn a remembered version. The distinction between the discount rate applied to the net liability (under IAS 19) and the expected return on assets (the older approach, still used in some regimes) is easily blurred, and remeasurement presentation is a detail candidates skip when revising.
What to do instead
Build a one-page reconciliation template you can reproduce from memory: opening net liability, service cost, net interest (discount rate times opening net liability, adjusted for contributions and benefit payments), remeasurements split into actuarial gains/losses and return on assets excluding interest, contributions paid, benefit payments, closing net liability. Practise classifying each item as P&L or OCI. Note the differences between IAS 19, FRS 102, and any local regime the syllabus covers, and state which standard you are applying at the start of the answer.
The candidate answers a risk-management question by listing risks (longevity, investment, inflation, sponsor, operational) without assessing their materiality, interaction, or the effectiveness of the proposed mitigation. The answer reads as a taxonomy rather than an analysis of this scheme's risk exposure.
Risk lists are easy to memorise and feel like content, so candidates reproduce them. The exam rewards prioritisation and judgement, and candidates are often unsure how to rank risks without a quantitative framework, so they avoid committing to a view.
What to do instead
For each risk, state the exposure in the scheme's own terms (e.g. a 1% fall in long-term rates increasing liabilities by roughly the duration times the liability value), then assess the direction and size of the impact on funding level and on the sponsor, then evaluate the specific mitigants available and their cost and residual risk. Conclude with a prioritised ranking and the reasoning for it. Where the question gives numbers, use them; where it does not, state an assumed order of magnitude and label it as an assumption.
In benefit-design or restructuring questions, the candidate proposes a change (e.g. closing the scheme to future accrual, moving to CARE, changing the retirement age) without addressing the legal and trustee constraints, the consultation requirements, the treatment of accrued rights, and the member-communication and industrial-relations consequences.
Candidates focus on the actuarial cost saving and treat implementation as a footnote. The professional and regulatory dimensions are examined heavily and are where marks are lost, because candidates do not know which protections are absolute (e.g. accrued rights generally cannot be reduced without member consent or a statutory process) and which are negotiable.
What to do instead
For any benefit-change question, run a checklist: what is the current benefit structure and which rights have already accrued; what statutory protection applies to accrued rights; what consultation or member-consent process is required; what are the cost, funding, and accounting consequences; what are the communication and industrial-relations risks; what transitional protections or grandfathering might apply. Answer in that order and quantify the cost impact of each option so the recommendation is evidence-based.
The candidate describes a decumulation or retirement-product question in generalities (annuity versus drawdown) without quantifying the trade-offs, addressing mortality and longevity risk pooling, the effect of the open market option, tax treatment, or the specific needs of the member cohort described.
Individual pension provision is often revised superficially as background to the occupational scheme material. Candidates do not practise the numerical comparison of products, and they default to describing product features rather than analysing suitability and value for the member.
What to do instead
For each option, quantify: the income at outset, the income at a stated later age, the impact of a stated investment return and inflation assumption, the value of any guarantees, and the tax treatment of the income and of any lump sum. Then match to member characteristics (health, dependants, other income, attitude to risk, need for flexibility). Discuss the role of shopping around and advice. Show the arithmetic rather than asserting that one option is 'more flexible'.
The candidate treats the trustee board and the sponsor as having identical interests, or fails to identify the conflicts of interest and the governance process that should manage them, when answering governance or ethical questions.
Candidates learn governance as a list of duties and committees but do not apply it to the conflicting objectives that arise in practice: funding negotiations, benefit changes, scheme wind-up, and the use of surplus. Ethical and professional questions are often answered with a statement of principle rather than an identification of the specific conflict and the process to manage it.
What to do instead
In every governance or ethics scenario, explicitly map the parties (members, sponsor, trustees, advisers), state each party's interest and where those interests diverge, identify the decision that is affected, and describe the process that should govern it: declaration and management of conflicts, independent advice, minutes and reasoning, member communication, and escalation to the regulator where appropriate. Where a professional code applies, name the specific principle engaged (e.g. integrity, competence, confidentiality) and how the proposed action satisfies it.
In wind-up and scheme-restructuring questions, the candidate allocates assets to liabilities without applying the statutory priority order and without distinguishing between the scheme's funding position on a technical provisions basis and on a buy-out or solvency basis.
Wind-up priority rules are detailed and jurisdiction-specific, and candidates avoid them because they are hard to remember precisely. They also conflate the ongoing funding target with the cost of securing benefits with an insurer, which can differ substantially and changes the answer entirely.
What to do instead
Learn the priority order for the jurisdiction in the syllabus (for UK schemes: the statutory debt, then pensions in payment and certain other protected liabilities, then the Pension Protection Fund assessment, then remaining benefits) and reproduce it as a numbered list at the start of the answer. Always compute or state both the ongoing funding position and the estimated buy-out cost, and explain why they differ. Then allocate the available assets down the priority order and identify the shortfall and who bears it.
The candidate answers a question about scheme design or state provision by describing the current system as if it were static, without addressing how the scheme or state benefit interacts with the member's other provision, means-tested benefits, or the tax system, and without considering distributional effects across member groups.
State pension and scheme-design material is often memorised as a set of rules rather than as an interacting system. Candidates do not practise tracing a member's total retirement income through the layers (state, occupational, personal), so they miss the interaction effects that carry marks.
What to do instead
For any question on individual or state provision, construct a worked example for a representative member: earnings history, accrual in each layer, retirement age, and resulting total income including any means-tested top-up. Then vary one assumption (retirement age, earnings, contribution rate) and describe the effect. This forces you to engage with interaction and distributional questions rather than reciting rules. Do the same for scheme-design comparisons: model the benefit for a short-service and a long-service member and compare.
The candidate runs out of time because they write long descriptive answers to the early questions and then give superficial answers to the later, higher-mark calculation or discussion questions, and they do not show the intermediate steps of calculations, so method marks are lost even where the final figure is close.
Exam technique is under-practised. Candidates revise content but do not rehearse under timed conditions, so they misjudge the marks-per-minute rate and the depth expected. They also write calculations as a single line, which makes it impossible for the marker to award method marks when an early slip occurs.
What to do instead
Before the exam, establish the marks-to-time ratio from past papers and allocate time per question, with a hard stop. In calculations, write each step on its own line with a label, state assumptions explicitly, and carry forward an obviously wrong figure rather than restarting, so downstream marks remain available. In discussion questions, use short headed paragraphs (one point per paragraph) rather than continuous prose. Practise at least two full papers under timed conditions and mark them against the scheme to calibrate depth.