Why Pool Multiple Experience Periods?
A pure premium is a loss cost per exposure unit. A single experience period may provide limited or unusually variable data, so a ratemaker may combine several periods to calculate one indication. The key is to bring the losses to a consistent prospective level, retain the exposures corresponding to each period, and pool the totals. The calculation uses the combined losses divided by the combined exposures—not an unadjusted average of the period results.
This tutorial focuses on that pooling calculation. It assumes that the exposure base has already been selected and that any needed loss development and trend selections are available. Earlier material on matching losses and exposures to the same basis remains important: pooling does not make incompatible data comparable.
The numerator and denominator must describe the same experience. If the losses for a period reflect one set of policies or coverage terms, its exposures must represent that same set. Losses also need to be adjusted using selections appropriate to the period and to the prospective level being indicated. For example, when ultimate losses and selected factors are supplied, a projected loss may be calculated as \(L_j^P=L_j^{ult}D_jT_j\), where \(D_j\) is a development factor and \(T_j\) is a trend factor. If losses are already at ultimate, do not apply development again.
Use a common definition of an exposure unit in every period—for example, one vehicle-year or one policy-year. The number of units can differ by period; that difference is handled by summing the exposures. The pooled result is a loss cost, not by itself a complete indicated rate. Other rate components, such as expenses or profit provisions, are outside this pooling calculation.
Pooling the Totals
A reliable workflow is to keep each period’s adjustment separate until its projected loss is calculated. Then add the projected losses, add the corresponding exposures, and divide. This prevents an adjustment intended for one period from being applied indiscriminately to the others.
Confirm that all periods use the same exposure unit and comparable coverage, and that each loss amount is paired with its own period’s exposures.
Apply the selected development and trend adjustments needed to place that period’s losses at the common prospective cost level. Do not reapply an adjustment already reflected in the input.
Calculate \(L_{\text{total}}^P=\sum_j L_j^P\) and \(E_{\text{total}}=\sum_j E_j\).
Calculate \(PP_{\text{pool}}=L_{\text{total}}^P/E_{\text{total}}\), then state the result in dollars per exposure unit.
The formula follows directly from the definition of a loss cost for the combined experience. Across all included periods, the total loss amount is the sum of the period losses, and the total number of exposure units is the sum of the period exposures. Therefore, the combined loss cost is total losses divided by total exposures. The period boundaries do not change that aggregate definition.
Worked Calculations
Worked Example: Pooling Two Periods at a Common Level
A hypothetical equipment insurer has projected losses of $390,000 for one experience period and $510,000 for a second period. The matching exposures are 12,000 and 18,000 equipment-years, respectively. Both loss amounts have already been adjusted to the same prospective level. Calculate the pooled pure premium.
Method. Sum projected losses and matching exposures separately, then divide. Since the amounts are already at the common level, no additional adjustment is needed.
Projected losses: \(390{,}000+510{,}000=900{,}000\) dollars.
Exposures: \(12{,}000+18{,}000=30{,}000\) equipment-years.
Thus,
The result is $30.00 per equipment-year. As a check, multiplying the pooled pure premium by the total exposures gives \(30.00\times30{,}000=\$900{,}000\), equal to the total projected losses. This calculation avoids treating the two periods as if they represented equal volumes of exposure.
Worked Example: Applying Period-Specific Development and Trend
A hypothetical specialty insurer has two periods of experience. Period 1 has reported losses of $300,000, a selected development factor of 1.10, a trend factor of 1.05, and 10,000 matching exposures. Period 2 has reported losses of $420,000, a development factor of 1.05, a trend factor of 1.10, and 15,000 matching exposures. The factors are selected to bring each period to the same prospective cost level. Calculate the pooled pure premium.
Method and assumptions. Treat the reported losses as not yet developed to ultimate. For each period, multiply reported losses by that period’s development and trend factors. The supplied factors are used as stated, and the exposure counts are on the same basis.
Period 1: Developed losses are \(300{,}000\times1.10=\$330{,}000\). After trend, projected losses are \(330{,}000\times1.05=\$346{,}500\).
Period 2: Developed losses are \(420{,}000\times1.05=\$441{,}000\). After trend, projected losses are \(441{,}000\times1.10=\$485{,}100\).
Totals: Projected losses are \(346{,}500+485{,}100=\$831{,}600\). Exposures are \(10{,}000+15{,}000=25{,}000\).
Therefore,
The pooled pure premium is $33.264 per exposure, or $33.26 per exposure if reported to the nearest cent. Keeping each period’s factors with its own losses avoids applying the wrong adjustment to a period.
Worked Example: Checking the Period Totals Before Dividing
A hypothetical rental-equipment portfolio has three periods. Its ultimate losses are $250,000, $360,000, and $500,000, and selected trend factors to the prospective level are 1.08, 1.05, and 1.04. The matching exposures are 8,000, 12,000, and 20,000 equipment-years. Calculate the pooled pure premium.
Method and assumptions. The losses are already ultimate, so apply trend but no further development. Use each period’s supplied trend factor and retain its corresponding exposure count.
Period 1: \(250{,}000\times1.08=\$270{,}000\).
Period 2: \(360{,}000\times1.05=\$378{,}000\).
Period 3: \(500{,}000\times1.04=\$520{,}000\).
Totals: Projected losses are \(270{,}000+378{,}000+520{,}000=\$1{,}168{,}000\). Exposures are \(8{,}000+12{,}000+20{,}000=40{,}000\) equipment-years.
The indication is
Check: \(40{,}000\times\$29.20=\$1{,}168{,}000\), matching the summed projected losses. The calculation also confirms that development was not applied a second time to losses already stated as ultimate.
Common Mistakes and Interpretation
A frequent error is to calculate each period’s pure premium and then take their simple arithmetic average. That treats every period as though it represented the same exposure volume. When exposure counts differ, it does not generally reproduce the combined losses per combined exposure. For the intended pooled indication, use the ratio of the two totals.
For instance, in the first worked example the period-level loss costs are \(390{,}000/12{,}000=\$32.50\) and \(510{,}000/18{,}000\approx\$28.33\). Their simple average is approximately $30.42, not the pooled result of $30.00. The simple average is the wrong calculation for the combined-experience loss cost; the correct calculation is \((390{,}000+510{,}000)/(12{,}000+18{,}000)\).
- Do not sum pure premiums. Sum projected loss dollars and exposure units, then divide.
- Do not mix exposure bases. A total made from unlike units has no coherent interpretation as a loss cost per exposure.
- Do not mispair periods. Each period’s projected losses must be divided against exposures corresponding to that experience.
- Do not double-adjust losses. Verify whether supplied losses are reported, ultimate, or already trended before applying factors.
- Do not mistake the result for a complete rate. The pooled pure premium is a loss cost per exposure; additional rate components may be needed in a full indication.
Pooling makes explicit how several periods contribute to one loss-cost measure. It is useful only when the periods have been prepared on a consistent basis and the selected adjustments are appropriate. A precise calculation cannot repair inconsistent coverage, exposure definitions, or loss treatment.
Check Your Understanding
For each problem, pool the experience by summing projected losses and exposures separately. Report the pure premium in dollars per exposure.
- A hypothetical property portfolio has projected losses of $280,000 and $360,000 for two periods, with 8,000 and 12,000 matching property-years. Find the pooled pure premium. (Answer: $32.00 per property-year)
- Three periods have projected losses of $155,000, $240,000, and $405,000 and matching exposures of 5,000, 10,000, and 15,000 policy-years. Find the pooled pure premium. (Answer: $26.67 per policy-year)
- A hypothetical insurer has ultimate losses of $200,000 and $300,000, trend factors of 1.10 and 1.05, and matching exposures of 5,000 and 10,000. Calculate each projected loss and then the pooled pure premium. (Answer: $35.67 per exposure)
- Two periods have projected losses of $480,000 and $520,000 and exposures of 12,000 and 8,000. A candidate takes the simple average of the two period pure premiums. What is the correct pooled pure premium? (Answer: $50.00 per exposure)
- For two periods, reported losses are $210,000 and $400,000, development factors are 1.00 and 1.00, trend factors are 1.20 and 1.05, and matching exposures are 7,000 and 13,000. Calculate the pooled pure premium. (Answer: $33.60 per exposure)