Converting Ultimate Losses into Prospective Pure Premium

From Developed Losses to a Prospective Cost

A historical loss amount is not yet a prospective pure premium. Losses may need to be developed to ultimate, trended to the future period being priced, and divided by the exposure units on the intended basis. The central distinction is that a loss ratio compares losses with premium, while a pure premium compares losses with exposure.

For this calculation, assume the ultimate losses have been estimated on a consistent coverage and claim basis, and that the exposure units represent the risks to which the indicated cost will apply. Trend the losses from the average date represented by each historical experience period to the average date of the prospective period. The trend factor and its time interval must be stated; the examples use a constant annual effective loss trend rate.

Definition: For experience periods $j=1,\ldots,m$, let $U_j$ be developed ultimate losses, $E_j$ be exposure units, and $n_j$ be the number of years from the average date of period $j$ to the prospective average date. With annual effective loss trend rate $t$, the prospective pure premium is $$ \frac{\sum_{j=1}^{m} U_j(1+t)^{n_j}}{\sum_{j=1}^{m} E_j}. $$ It is the projected loss cost per exposure unit. The formula assumes the exposure units are comparable across periods and that the stated trend appropriately adjusts losses to the prospective basis.

When a reported or paid amount is provided instead of an ultimate loss, first use the supplied development estimate or factor to obtain ultimate losses. Development and trend address different changes: development estimates the eventual amount of losses from a historical period; trend adjusts that ultimate amount for changes in the cost level between the historical and prospective dates. Applying trend does not replace development.

Why the Calculation Works

For one experience period, the ultimate loss cost on its historical basis is $U/E$. If the expected loss amount per exposure unit changes by the annual factor $1+t$, then after $n$ years the projected cost is $U/E$ multiplied by $(1+t)^n$. Multiplying the cost per exposure unit by the period's exposure gives its projected losses:

E\left(\frac{U}{E}(1+t)^n\right)=U(1+t)^n.

For several periods, projected losses add across periods, as do exposure units. Dividing the sum of projected losses by the sum of exposures therefore produces the combined prospective loss cost per exposure unit. This also shows why an unweighted average of the periods' loss costs is generally wrong: it gives the same weight to periods with different exposure volumes.

Key distinction: A historical loss ratio is ultimate losses divided by earned premium. An on-level loss ratio uses premium adjusted to a common rate level in the denominator. A pure premium is projected losses divided by exposure. On-level premium is useful for expressing losses relative to a common premium level, but it is not itself an exposure denominator.

If the available premium has been adjusted to the prospective rate level, a consistent loss-ratio calculation can also produce the pure premium. For a single period, divide trended ultimate losses by on-level premium to get the trended loss ratio, then multiply by on-level premium per exposure unit:

\left(\frac{U(1+t)^n}{P_{\mathrm{on}}}\right) \left(\frac{P_{\mathrm{on}}}{E}\right) = \frac{U(1+t)^n}{E},

where $P_{\mathrm{on}}$ is earned premium adjusted to the chosen common rate level. The equality is an accounting identity: the premium cancels. It is valid only when the loss ratio and premium per exposure use compatible premium, exposure, coverage, and time bases.

Worked Applications

Worked Example: Develop, Trend, and Divide by Exposure

A hypothetical insurer has $360,000 of reported losses for a completed period. Its selected ultimate-to-reported factor is 1.25. The period represents 1,000 exposure units, and the average date of the period is two years before the prospective average date. Assume annual effective loss trend of 5%. Calculate the prospective pure premium.

Method. Develop the reported losses to ultimate, trend that ultimate amount, then divide by exposure.

Step 1: Develop to ultimate. $$ U=360{,}000(1.25)=450{,}000. $$ The $450,000 is the estimated ultimate loss for the historical period, before prospective trend.

Step 2: Calculate the trend factor. There are two years between the stated average dates: $$ (1.05)^2=1.1025. $$

Step 3: Project losses to the prospective date. $$ 450{,}000(1.1025)=496{,}125. $$

Step 4: Divide by the matching exposure. $$ \frac{496{,}125}{1{,}000}=\$496.125. $$ Rounded to the nearest cent, the indicated prospective pure premium is $496.13 per exposure unit.

Check. The historical ultimate loss cost is $450,000 / 1,000 = $450.00 per exposure unit. Applying the two-year factor gives $450.00 × 1.1025 = $496.125, consistent with the direct calculation. This method avoids the mistake of applying development and trend as if they were the same adjustment.

Worked Example: Combining Two Experience Periods

For a hypothetical coverage, developed ultimate losses and earned exposures are $300,000 and 1,000 units for the first period, and $420,000 and 1,200 units for the second period. The first period's average date is two years before the prospective average date; the second is one year before it. Assume 5% annual effective loss trend. Find the combined prospective pure premium.

Method. Trend each period's ultimate losses for its own interval, then sum projected losses and exposures.

First period. Its trend factor is $$ (1.05)^2=1.1025. $$ Projected losses are $$ 300{,}000(1.1025)=330{,}750. $$

Second period. Its trend factor is $$ (1.05)^1=1.05. $$ Projected losses are $$ 420{,}000(1.05)=441{,}000. $$

Combine periods. Total projected losses are $330,750 + $441,000 = $771,750. Total exposures are 1,000 + 1,200 = 2,200. Thus, $$ \frac{771{,}750}{2{,}200}=\$350.79545\ldots $$ Rounded to the nearest cent, the indicated prospective pure premium is $350.80 per exposure unit.

Check. The first period contributes $330,750 / 2,200 = $150.3409 per combined exposure unit, and the second contributes $441,000 / 2,200 = $200.4545. Their sum is $350.7954. This check confirms that both periods' projected losses are included and that the final denominator is total exposure.

Common mistake avoided. Averaging the two period-specific loss costs equally would give $(330.75+367.50)/2=\$349.125$. That is not the combined indication because the periods have different exposure volumes. The correct calculation weights each period through its losses and exposure.

Worked Example: Reconciling On-Level Loss Ratios and Pure Premium

Two hypothetical experience periods have the following data. Period A has $200,000 of developed ultimate losses, 500 exposure units, $300,000 of earned premium, and an on-level premium factor of 1.20. Its average date is two years before the prospective average date. Period B has $330,000 of developed ultimate losses, 600 exposure units, $450,000 of earned premium, and an on-level factor of 1.10; its average date is one year before the prospective date. Assume annual effective loss trend of 10%. Calculate the prospective pure premium, and explain why the historical loss ratios alone do not give it.

Method. First adjust each premium to the common rate level. Then trend the losses by period and compare their total with total exposure. The on-level premiums provide a cross-check through a trended on-level loss ratio.

Step 1: Calculate on-level premiums. $$ P_{\mathrm{on},A}=300{,}000(1.20)=360{,}000, \qquad P_{\mathrm{on},B}=450{,}000(1.10)=495{,}000. $$ Total on-level premium is $360,000 + $495,000 = $855,000.

Step 2: Trend losses to the prospective date. Period A has factor $(1.10)^2=1.21$, so its projected losses are $$ 200{,}000(1.21)=242{,}000. $$ Period B has factor $(1.10)^1=1.10$, so its projected losses are $$ 330{,}000(1.10)=363{,}000. $$ Total projected losses are $242,000 + $363,000 = $605,000.

Step 3: Calculate the pure premium directly. Total exposure is 500 + 600 = 1,100 units, giving $$ \frac{605{,}000}{1{,}100}=\$550.00 $$ per exposure unit.

Step 4: Reconcile using on-level premium. The trended on-level loss ratio is $605,000 / $855,000 = 0.707602\ldots. On-level premium per exposure is $855,000 / 1,100 = $777.2727\ldots. Their product is $$ \left(\frac{605{,}000}{855{,}000}\right) \left(\frac{855{,}000}{1{,}100}\right) = \frac{605{,}000}{1{,}100} = \$550.00. $$

Check and interpretation. The historical loss ratios using actual earned premium are $200,000 / $300,000 = 66.67% for Period A and $330,000 / $450,000 = 73.33% for Period B. These ratios reflect the historical premium levels, not the common on-level premium basis, and neither includes the required trend adjustment. The reconciled calculation agrees with the direct pure premium.

Common Traps and Basis Checks

Before reporting an indication, identify the amount being projected, the date to which it is projected, and the units in the denominator. The following checks catch frequent errors:

Using reported losses as ultimate losses. If development is required, apply the stated ultimate estimate before trending.
Using one trend interval for every period. Calculate the time from each historical period's average date to the prospective average date. A one-year error changes the trend factor.
Dividing by premium when asked for a pure premium. Premium belongs in a loss ratio; exposure belongs in a cost-per-exposure calculation.
Using actual premium where on-level premium is required. A historical loss ratio based on a different rate level is not an on-level loss ratio.
Averaging period indications without weights. For a combined indication, add projected losses and exposures before dividing.
Mixing incompatible bases. Losses, premiums, and exposures must correspond to consistent coverage, limits, deductibles, and time periods. Trend should not be used to conceal a change in the underlying exposure definition.

For example, in the preceding two-period case, it would be incorrect to calculate the historical loss ratio as $530,000 / $750,000 and multiply it by on-level premium per exposure. The numerator is not trended, and the denominator is not on-level. A valid loss-ratio route uses projected losses over on-level premium; the premium-per-exposure factor then converts that ratio to a pure premium. The direct route, projected losses divided by exposure, makes the required denominator especially clear.

Key takeaway: Develop losses to ultimate when needed, trend each period to the prospective date, and divide total projected losses by the corresponding total exposure. Use on-level premium for a rate-level-consistent loss ratio, not as a substitute for exposure in a pure-premium calculation.

Check Your Understanding

For each question, use the stated trend interval and round dollar pure premiums to the nearest cent.

A hypothetical book has $275,000 of developed ultimate losses, 1,000 exposure units, and a two-year interval to the prospective average date. At 4% annual effective loss trend, what is the prospective pure premium? (Answer: $297.44 per exposure unit)
Two periods have ultimate losses of $180,000 and $240,000, with 600 and 800 exposure units, respectively. Their trend intervals are two years and one year. At 3% annual effective trend, what is the combined prospective pure premium? (Answer: $312.97 per exposure unit)
A period has $300,000 of ultimate losses, $500,000 of actual earned premium, $575,000 of on-level premium, and 1,000 exposure units. Its prospective trend interval is one year at 8%. Find the prospective pure premium and the trended on-level loss ratio. (Answer: $324.00 per exposure unit; trended on-level loss ratio is 56.35%)
A period has $320,000 of reported losses, an ultimate-to-reported factor of 1.25, 800 exposure units, and a three-year interval to the prospective average date. At 5% annual effective loss trend, find the prospective pure premium. (Answer: $578.81 per exposure unit)
Two periods have ultimate losses of $250,000 and $360,000 and exposures of 500 and 900 units. Their trend intervals are two years and one year, respectively. At 6% annual effective trend, calculate the combined prospective pure premium. (Answer: $473.21 per exposure unit)

What You'll Learn