Expected Loss Ratio Method

Insurance losses develop over time: a recently written period may have few reported claims even though more will emerge or existing claims will become more expensive, and an expected loss ratio supplies a prospective basis for estimating ultimate cost. The ratio compares expected losses with the relevant earned premium, and it should use a consistent definition of losses and premium, because mixing gross losses with net premium, or including expenses on only one side, changes what the estimate represents.

Earned premium is the portion associated with coverage provided during the measured period, which differs from simply counting all cash collected or all premium written, so align the exposure period before multiplying by the ratio. The selected ratio can draw on pricing assumptions, historical experience and relevant industry evidence, but changes in business mix, rates, claims inflation and coverage affect whether that evidence remains useful.

A previous year's ratio is not automatically appropriate for the current portfolio. Segmenting the portfolio can improve relevance, since different lines or cohorts may have different loss expectations and development speeds, and an aggregate ratio can conceal a shift toward higher-risk business even when the total premium base is clear.

Ultimate losses include amounts already paid and amounts expected to remain unpaid, so the method's product is an ultimate estimate, not necessarily the reserve balance. Subtracting the appropriate paid losses is required when estimating the remaining unpaid amount on a consistent basis.

A claims estimate can use reported rather than paid information for other purposes, and if deriving unreported amounts the relevant subtraction differs from estimating total unpaid claims, so label which reserve or component is being calculated instead of calling every difference IBNR. The CAS paper Using Expected Loss Ratios in Reserving explores expected ratios alongside loss-development information and credibility, including approaches that blend expected losses with reported experience.

That broader analysis illustrates why selecting and evaluating assumptions matters, rather than implying every method uses only one multiplication. The Bornhuetter-Ferguson approach is distinct from the simple expected loss ratio method because it combines observed losses with an expectation for the remaining unreported or unpaid portion under the selected formulation, so do not apply its formula while labelling the result a pure expected-ratio estimate.

Loss-development methods rely more directly on emerging claims and development patterns, and they can differ materially from an expected-ratio estimate when early experience is volatile. Comparing methods can reveal assumption sensitivity without automatically proving one result is correct.

Scenario analysis makes uncertainty visible: recalculate ultimate and unpaid losses under reasonable alternative ratios and explain why those alternatives were selected, remembering that a range shows dependence on assumptions, not a verified probability distribution by itself. For a non-finance manager, ask which premium and loss definitions are used, how the ratio was selected and whether conditions have changed.

Reconcile the ultimate estimate with paid and reported claims, and compare it with emerging evidence. Treat the result as an assumption-based estimate needing review, not a final invoice or guaranteed claim total.