AIAF common mistakes
AIAF · Understand the traps and how to avoid them.
Applies GAAP matching and accrual logic to statutory-basis questions - for example capitalizing deferred acquisition costs or recording non-admitted assets at full value when the item asks for statutory surplus.
Most candidates learn GAAP accounting first and default to it unconsciously; statutory accounting's conservatism (non-admitted assets, immediate DAC expensing, prescribed reserve levels) feels like the exception rather than the rule being tested.
What to do instead
Before answering, write 'SAP' or 'GAAP' at the top of the scratch space and force every adjustment through a two-column checklist: (1) Is this asset admitted under statutory rules? (2) Is this liability recorded at the prescribed conservative amount? Practice by taking one balance sheet and re-stating it both ways until the differences become automatic.
Mixes premium streams - reporting written premium when the question asks for earned, or netting ceded premium into direct premium before computing a ratio.
The words 'written', 'earned', 'direct', 'assumed', 'ceded' and 'net' appear in dense stems and the candidate latches onto the first number rather than mapping each term to its definition.
What to do instead
Build a one-line premium bridge from memory (direct written + assumed written = gross; gross − ceded = net written; net written ± unearned change = net earned) and physically label every figure in the stem with its bridge position before touching a calculator.
Treats IBNR and case reserves as substitutes rather than components, or applies loss development to case reserves only, producing a materially understated loss reserve and loss ratio.
Case reserves are the visible, named amounts in a problem, so they attract attention; IBNR is an estimate that requires an explicit development or expected-loss step the candidate skips under time pressure.
What to do instead
Always compute total reserve as case + IBNR and show the development factor applied to reported losses explicitly; when a question supplies a development triangle or a percentage of earned premium for IBNR, state which method is being used and why before calculating.
Reads the combined ratio in isolation and concludes the insurer is unprofitable, ignoring investment income, or reverses the components (puts the expense ratio where the loss ratio belongs).
The combined ratio is the most familiar number on the page, so candidates anchor on it and stop analysis; the components look similar when written as decimals and get transposed.
What to do instead
Recite the component order as a formula first - loss ratio + expense ratio + policyholder dividend ratio = combined ratio - then compute the operating ratio by subtracting the investment income ratio, and state in one sentence what each component measures before using it in a conclusion.
Adds reinsurance recoverables to surplus at full carrying value without deducting the allowance for uncollectible reinsurance, or classifies a surplus note as a liability/equity without checking the statutory treatment.
Recoverables sit in the asset section and look like cash; surplus notes have debt-like features (interest, maturity) that trigger a GAAP instinct even though statutory treatment differs.
What to do instead
Memorize the short list of statutory items with special treatment (reinsurance recoverables and their allowance, surplus notes, non-admitted assets, EDP equipment) and, for each, write the one-line rule and its surplus effect before answering any balance-sheet question.
Computes duration or asset-liability matching using book value and the stated maturity, ignoring market value, embedded options, and policyholder lapse or surrender behavior on the liability side.
Duration formulas are usually practiced on clean fixed-coupon examples, so the candidate reuses the textbook mechanics and never adjusts for the call features in the bond portfolio or the optionality in the liability.
What to do instead
For every ALM question, list three inputs explicitly: valuation basis (market value), cash-flow timing including calls/prepays, and liability behavior assumptions. If a question supplies a liability duration or lapse assumption, incorporate it rather than substituting the asset duration.
Solves capital-adequacy questions by summing risk charges arithmetically instead of applying the correlation/covariance adjustment, or by comparing the wrong capital base to the company action level.
The risk-based capital formula is a multi-step calculation and candidates collapse it into a single addition to save time; the authorized-control-level versus company-action-level thresholds blur under exam pressure.
What to do instead
Write the RBC skeleton (five risk categories → covariance adjustment → total RBC → compare to ACL/regulatory-action levels) as a reusable template and plug numbers into the template rather than into a shortcut; check the threshold labels against the ratio computed.
Confuses the statutory surplus measures used for dividend and growth tests - for example using total surplus when the question requires surplus as regards policyholders, or computing dividend capacity from net income rather than earned surplus.
Several surplus and dividend definitions differ by only a word and appear together in regulatory questions; candidates recognize the term but not the distinguishing modifier.
What to do instead
Maintain a definition card for each regulatory measure (surplus as regards policyholders, earned surplus, free surplus, dividend capacity) with the exact formula and one numerical mini-example, and reread the stem to confirm which measure is named before calculating.
Answers ethics and professional-conduct items with a gut reaction - 'that's obviously wrong' - instead of identifying the specific duty breached, the affected parties, and the required escalation or disclosure step.
Ethics content feels like common sense, so candidates spend little time on it and lose points on questions that turn on the order of actions or on disclosure timing rather than on whether the conduct was bad.
What to do instead
Use a fixed four-step frame for every ethics item: identify the duty or standard implicated, name who is harmed, choose the action that preserves the reporting chain (escalate internally first, document, then external disclosure if unresolved), and only then select the option that fits that sequence.
Loses points on multi-step computational items because units and bases are never reconciled - mixing thousands and millions, applying an annualized ratio to a quarterly figure, or applying a percentage to the wrong denominator.
Exam items are written with realistic financial statements where scale changes between exhibits and stems, and the candidate moves straight to the formula without normalizing the inputs.
What to do instead
Adopt a two-line pre-calculation ritual: restate every input in the same units and the same time basis (annual vs quarterly), and write the denominator of every ratio in words before dividing. If a final answer differs from all options by a factor of four or a thousand, that is the signal to recheck scale, not to pick the closest number.