risk adjustment

Suppose someone offers to take over your uncertain future bills in exchange for a fee. The pure expected value of those bills is one number — but you'd happily pay a bit more than the average to be rid of the uncertainty itself, and the person taking it on would demand that extra to compensate them for bearing it. That 'extra for bearing uncertainty' is what the IFRS 17 risk adjustment captures.

Formally, the risk adjustment for non-financial risk is the compensation an insurer requires for bearing the uncertainty in the amount and timing of cash flows arising from non-financial risks — chiefly insurance risk such as mortality, lapse, and expense, but not financial risk like interest rates (which is handled through discounting). It sits on top of the present value of expected cash flows in the measurement of the insurance liability, making the liability prudently larger than the bare expected value. IFRS 17 does not prescribe a single method; common approaches include a cost-of-capital calculation or a confidence-level (value-at-risk) technique, and the standard requires the insurer to disclose the confidence level its risk adjustment corresponds to so readers can compare. For example, expected claims of 1,000 might carry a risk adjustment of 70, giving a liability building block of 1,070 before the CSM.

Why it matters: the risk adjustment is where actuarial judgment about the genuine riskiness of a book of business becomes a hard number on the balance sheet, and as risk runs off over time it is released into profit. A common misconception is that it is an arbitrary safety pad — it is meant to be an entity-specific, justifiable measure of how much that particular insurer values being relieved of uncertainty, and it must be disclosed and defended.