Applying Loss Trend to Pure Premium Indications

Why Trend Loss Costs?

A pure premium is expected loss per exposure unit. Historical losses may be measured at an earlier cost level than the future period for which a rate is being indicated. If the selected loss trend represents the expected change in loss cost over time, applying it brings historical experience to the intended future cost level. The key is to apply the trend to losses or loss costs while keeping the exposure denominator on its intended basis.

This tutorial takes the selected trend rate and trend interval as inputs. The interval is the time from the experience loss-cost level to the target loss-cost level; choosing that interval is a separate judgment. The experience losses and exposures used below are assumed to be on a consistent basis already, including any required loss development, coverage, and premium-level adjustments.

Definition: If the selected effective annual loss trend rate is $r$, and the loss-cost level is projected forward by $n$ years, the compounded trend factor is $T=(1+r)^n$. For historical losses $L$ and corresponding exposures $E$, the historical pure premium is $L/E$, and its trended pure premium is $(L/E)T$. The trend changes the loss cost per exposure; it does not, by itself, change the number of exposures.

For a fractional year, $n$ may be fractional. Use the compounding convention that matches the selected trend rate. For example, if a rate is an effective annual rate and the trend interval is one-half year, the factor is $(1+r)^{0.5}$, not $1+r/2$. The latter is a simple proportional approximation, not the stated effective annual compounding model.

Applying Trend to One Experience Period

For one period, first calculate loss cost on the experience exposure basis. Then apply the trend factor. If a future exposure volume is also supplied, use it only after obtaining the trended loss cost: multiplying that cost by future exposures estimates total future losses.

Worked Example: Trend One Period and Estimate Future Losses

A hypothetical program has $960,000 of consistent experience losses and 6,000 earned exposure units. The selected effective annual loss trend is 6%, and the target loss-cost level is two years after the experience level. The projected target-period exposure volume is 8,500 units. Find the trended pure premium and the corresponding projected total losses.

Method. Calculate the experience pure premium, compound the selected trend for two years, and then multiply the trended pure premium by the target-period exposures.

Step 1: Historical pure premium.

\frac{L}{E}=\frac{\$960{,}000}{6{,}000}=\$160.00 \text{ per exposure}

Step 2: Trend factor. The rate is effective annual and the interval is two years, so

$$ T=(1.06)^2=1.1236 $$

Step 3: Trended pure premium.

\$160.00(1.1236)=\$179.776 \text{ per exposure}

Retaining precision until the end, this is $179.78 per exposure when rounded to cents.

Step 4: Projected total losses. Apply the trended cost to the 8,500 target-period exposures:

\$179.776(8{,}500)=\$1{,}528{,}096

Check. The historical losses trended at the original 6,000-unit exposure volume are $\$960{,}000(1.1236)=\$1{,}078{,}656$. Dividing by 6,000 gives $179.776 per exposure, matching Step 3. Scaling that cost to 8,500 exposures gives the projected total above. The common mistake avoided here is treating a change in exposure volume as if it were part of the loss trend.

Pooling Periods at a Common Target Level

When experience spans multiple periods, each period may be at a different loss-cost level. Trend each period’s losses using its own interval to the common target. Then add the trended losses and divide by the corresponding total experience exposures. This is equivalent to taking an exposure-weighted average of the individually trended pure premiums.

Worked Example: Pool Two Periods Before Calculating the Pure Premium

A hypothetical insurer has two annual experience periods. The first has $600,000 of losses and 4,000 exposures; its loss-cost level is two years before the target. The second has $840,000 of losses and 5,000 exposures; its level is one year before the target. The selected effective annual loss trend is 5%. Find the pooled pure premium at the target level.

Method. Trend each period’s losses separately to the same target, then pool the trended losses over the 9,000 corresponding exposures.

Step 1: Calculate the period-specific factors.

T_1=(1.05)^2=1.1025,\qquad T_2=(1.05)^1=1.05

Step 2: Trend each period’s losses.

L_1^t=\$600{,}000(1.1025)=\$661{,}500

L_2^t=\$840{,}000(1.05)=\$882{,}000

Step 3: Pool the trended losses and exposures.

L^t_{\text{total}}=\$661{,}500+\$882{,}000=\$1{,}543{,}500

E_{\text{total}}=4{,}000+5{,}000=9{,}000

Step 4: Calculate the pooled pure premium.

\frac{\$1{,}543{,}500}{9{,}000}=\$171.50 \text{ per exposure}

Check. The individual trended pure premiums are $\$661{,}500/4{,}000=\$165.375$ and $\$882{,}000/5{,}000=\$176.40$. Weighting them by their respective exposures gives $(4{,}000(\$165.375)+5{,}000(\$176.40))/9{,}000=\$171.50$. A simple average of those two period costs would be $170.8875, which is not the pooled pure premium because the periods have different exposure volumes.

Why the Formula Preserves the Exposure Basis

The formula follows from separating a per-exposure cost from the number of exposures. Let the experience pure premium be $q=L/E$. Under the selected trend model, the target-level expected loss cost per exposure is $qT$. For a target volume of $E_f$ exposures, the corresponding total expected losses are $E_f qT$. Dividing those losses by $E_f$ returns the trended pure premium:

\frac{E_f qT}{E_f}=qT=\frac{L}{E}T

This also explains why projected exposure volume does not belong inside the trend factor. Exposure volume determines the total expected losses after the per-exposure cost has been projected. It does not change the cost per exposure unless a separate change in exposure composition or risk characteristics is being modeled.

Result: For experience periods $j=1,\ldots,k$, let $L_j$ be losses, $E_j$ be corresponding exposures, and $T_j$ be the factor from period $j$ to one common target level. The pooled trended pure premium is $$ \frac{\sum_{j=1}^{k} L_jT_j}{\sum_{j=1}^{k}E_j}. $$ This calculation assumes the losses and exposures are comparable across periods and that the selected trends apply to the loss costs being measured. It does not correct for a changing exposure mix or inconsistent coverage by itself.

Compound Trend, Total Losses, and a Time-Unit Trap

A stated annual effective trend compounds over successive years. A multi-year interval therefore requires repeated multiplication, represented by an exponent. Using a simple annual percentage multiplied by the number of years understates the compounded factor when the trend rate is positive. Also keep the outputs distinct: the trended pure premium is a cost per exposure, while projected total losses additionally depend on the target exposure count.

Worked Example: Four-Year Compounding on a Fixed Exposure Unit

A hypothetical portfolio has $1,440,000 of losses over 12,000 exposure units, already on a consistent basis. The selected effective annual loss trend is 5%, and the target loss-cost level is four years later. The projected target volume is 13,500 units. Calculate the target pure premium and projected total losses.

Method. Calculate the observed cost per unit, compound the annual trend for four years, and then apply the trended cost to the projected volume.

Step 1: Observed pure premium.

\frac{\$1{,}440{,}000}{12{,}000}=\$120.00

Step 2: Four-year trend factor.

$$ T=(1.05)^4=1.21550625 $$

Step 3: Target pure premium.

\$120.00(1.21550625)=\$145.86075

The target pure premium is $145.86 per exposure when rounded to cents.

Step 4: Target total losses.

\$145.86075(13{,}500)=\$1{,}969{,}120.125

Rounded to cents, projected total losses are $1,969,120.13.

Check and trap. Using the linear shortcut $1+4(0.05)=1.20$ would give a pure premium of $144.00, not $145.86075. The selected model is annual compounding, so the correct factor is $(1.05)^4$. The total-loss calculation also uses the projected 13,500 exposures, not the historical 12,000.

Assumptions and Common Errors

A trend calculation is only meaningful if its inputs describe the same intended quantity. A selected loss trend should apply to the type of losses being used, the trend interval should match the stated target level, and the loss and exposure measures should be compatible. If these conditions do not hold, arithmetic can be correct while the indication is not.

Trending exposure counts. A loss trend adjusts loss cost, not exposure volume. Project exposure separately when a future total is required.
Using one factor for periods with different intervals. If the periods have different time distances to the target, apply their respective factors before pooling.
Taking a simple average of period pure premiums. Pool by dividing total trended losses by total matching exposures; this weights each period by its exposure volume.
Using the wrong time unit. Convert the interval to years when the selected rate is effective annual. Do not silently treat a monthly or quarterly interval as a full year.
Applying trend twice. If losses have already been adjusted to the target trend level, do not apply the same trend factor again.
Ignoring comparability. Trend does not automatically reconcile different exposure definitions, coverages, or mixes. Those require separate treatment rather than being hidden in the trend factor.

Applying loss trend correctly preserves the meaning of a pure premium: expected losses per stated exposure unit at a specified cost level. That distinction is useful in a rate indication because it separates changes in expected cost from changes in the amount of business being written. The treatment here is aligned with the CAS Exam 5 (Basic Ratemaking and Estimated Claim Liabilities) syllabus.

Check Your Understanding

For each problem, treat the stated losses as already comparable with the stated exposures. Retain precision during calculations and round final dollar amounts to cents.

A hypothetical class has $525,000 of losses over 3,500 exposures. The selected effective annual trend is 4% for two years. What is the trended pure premium? (Answer: $162.24 per exposure.)
Two experience periods have losses and exposures of $450,000 and 3,000, and $720,000 and 4,000. Their respective intervals to the target are two years and one year. At a 3% effective annual trend, calculate the pooled target pure premium. (Answer: $174.14 per exposure.)
A historical pure premium is $92 per exposure. Apply a selected 6% effective annual trend for three years, then estimate total losses for a target volume of 12,000 exposures. (Answer: $109.57 per exposure; $1,314,881.66 total losses.)
A hypothetical program has $1,020,000 of losses over 6,000 exposures. Trend is 7% effective annually for two years, and projected target volume is 7,500 exposures. Find the trended pure premium and projected total losses. (Answer: $194.63 per exposure; $1,459,747.50 total losses.)
One period has a pure premium of $100 over 8,000 exposures; another has a pure premium of $140 over 2,000 exposures. Both are one year before the target, and the selected effective annual trend is 10%. Find the pooled target pure premium. (Answer: $118.80 per exposure.)

What You'll Learn