Deriving the Combined Adjusted Loss Ratio

Why the Combined Adjusted Loss Ratio Is Needed

An observed loss ratio compares losses with the premium earned during the experience period. For a rate indication, however, the two sides of that ratio may need different adjustments. Losses may need to be developed to ultimate and brought to a prospective cost level; premiums may need to be restated at the current rate level. The resulting ratio compares losses and premium on consistent bases.

The key is to keep track of what each factor changes. A development factor and a trend factor adjust losses. An on-level factor adjusts premium. Combining the factors in the wrong place, or using one adjustment as if it applied to both sides, changes the meaning of the calculation.

Definition: For an experience period, the combined adjusted loss ratio is adjusted losses divided by on-level premium. Adjusted losses are losses brought to the selected ultimate basis and then to the selected prospective cost level. On-level premium is earned premium restated at the current rate level. The ratio is commonly expressed as a percentage.

Deriving the Formula

Let $L$ be the observed losses for an experience period and $P$ its earned premium. Let $D$ be the selected loss development factor, $T$ the selected loss trend factor, and $O$ the premium on-level factor. In this tutorial, each factor is supplied on a consistent basis: $D$ converts the observed losses to the selected ultimate basis, $T$ brings those losses to the prospective cost level, and $O$ restates premium to the current rate level.

The adjustments follow from their definitions. First, the selected ultimate losses are $LD$. Applying the trend factor to those losses gives $LDT$. Separately, the on-level premium is $PO$. Therefore, the ratio of adjusted losses to on-level premium is:

\text{Combined adjusted loss ratio} = \frac{LDT}{PO}.

This expression also has a useful ratio-of-adjustments form. Since the observed loss ratio is $L/P$, dividing the adjusted losses by on-level premium gives:

\frac{LDT}{PO} = \frac{L}{P}\times\frac{DT}{O}.

This derivation works because the numerator is multiplied by the factors that transform losses, while the denominator is multiplied by the factor that transforms premium. It does not mean that every factor should be applied to the observed loss ratio in the same direction. The premium factor is in the denominator because on-leveling increases or decreases the premium used for comparison.

Assumptions and scope: The factors are multiplicative and are appropriate to the same experience data and target basis. The trend factor is selected for losses on the chosen development basis, and the on-level factor converts the relevant earned premium to the current rate level. Do not apply a development factor again to losses already selected at ultimate, or include a trend factor that has already been incorporated into the selected losses. This deterministic calculation does not require an interest rate, mortality assumption, or probability distribution.

When all observations share the same factors, the ratio-of-adjustments form is convenient. When different groups have different factors, calculate adjusted losses and on-level premiums group by group, then add each side. A single aggregate loss ratio multiplied by simple averages of the factors is generally not equivalent: the appropriate weights differ between losses and premiums.

Applying the Adjustments in Sequence

Identify the loss basis.
Determine whether the losses are reported, incurred, or already selected at ultimate. Use a development factor only to make the required development adjustment.

Adjust the losses.
Multiply the applicable losses by the selected development and trend factors. These factors belong on the loss side of the calculation.

Adjust the premium.
Multiply earned premium by its applicable on-level factor. This produces premium on the current rate level.

Form and interpret the ratio.
Divide adjusted losses by on-level premium. The resulting percentage is the adjusted loss ratio on the specified bases; it is not, by itself, a statement of the direction or size of a rate change.

Worked Calculations

Worked Example: One Experience Period With Three Adjustments

A hypothetical commercial equipment program has $ \$480{,}000 $ of observed losses and $ \$900{,}000 $ of earned premium. The selected development factor is 1.10, the trend factor is 1.04, and the premium on-level factor is 1.05. Find the combined adjusted loss ratio. All amounts are in dollars. No interest, mortality, or probability-distribution assumptions are used; the supplied factors are deterministic.

Method. Adjust losses by development and trend, adjust premium by the on-level factor, and divide the adjusted losses by the on-level premium.

Step 1: Develop the losses.

\$480{,}000(1.10)=\$528{,}000.

Step 2: Trend the developed losses.

\$528{,}000(1.04)=\$549{,}120.

Step 3: On-level the premium.

\$900{,}000(1.05)=\$945{,}000.

Step 4: Divide adjusted losses by on-level premium.

\frac{\$549{,}120}{\$945{,}000} =0.581079\ldots =58.11\%.

Check. The observed loss ratio is $ \$480{,}000/\$900{,}000=0.533333\ldots $. Multiplying it by $1.10(1.04)/1.05$ gives $0.581079\ldots$, matching the direct calculation. The answer is rounded to two decimal places as a percentage.

Worked Example: Aggregating Groups With Different Factors

A hypothetical marine cargo portfolio is split into two rating groups. Group A has $ \$300{,}000 $ of observed losses, a development factor of 1.20, a trend factor of 1.05, $ \$600{,}000 $ of earned premium, and an on-level factor of 1.04. Group B has $ \$400{,}000 $ of observed losses, a development factor of 1.05, a trend factor of 1.02, $ \$800{,}000 $ of earned premium, and an on-level factor of 1.08. Find the aggregate combined adjusted loss ratio. Amounts are in dollars; there are no interest, mortality, or distribution assumptions.

Method. Because the groups have different factors, adjust each group's losses and premium separately before aggregating.

Group A losses.

\$300{,}000(1.20)(1.05) =\$360{,}000(1.05) =\$378{,}000.

Group B losses.

\$400{,}000(1.05)(1.02) =\$420{,}000(1.02) =\$428{,}400.

Total adjusted losses are $ \$378{,}000+\$428{,}400=\$806{,}400 $.

Group A and B on-level premiums.

\begin{aligned} P_{A,\mathrm{on\text{-}level}}&=\$600{,}000(1.04)=\$624{,}000,\\ P_{B,\mathrm{on\text{-}level}}&=\$800{,}000(1.08)=\$864{,}000. \end{aligned}

Total on-level premium is $ \$624{,}000+\$864{,}000=\$1{,}488{,}000 $. Thus:

\frac{\$806{,}400}{\$1{,}488{,}000} =0.541935\ldots =54.19\%.

Check. The numerator and denominator each equal the sum of the relevant group-level adjusted amounts. The result is rounded to two decimal places as a percentage. Averaging the four factors first and applying those averages to total losses or premium would not generally preserve these group-level totals.

Worked Example: Avoiding Development of Ultimate Losses Twice

For a hypothetical specialty liability portfolio, one segment has selected ultimate losses of $ \$360{,}000 $, a trend factor of 1.05, and earned premium of $ \$400{,}000 $ with an on-level factor of 1.08. A second segment has observed losses of $ \$240{,}000 $, a development factor of 1.25, a trend factor of 1.03, and earned premium of $ \$300{,}000 $ with an on-level factor of 1.02. Find the combined adjusted loss ratio. Amounts are in dollars; no interest, mortality, or distribution assumptions are used.

Method. The first segment's losses are already ultimate, so no additional development factor is applied. The second segment's observed losses still require the stated development factor.

Segment 1 adjusted losses.

\$360{,}000(1.05)=\$378{,}000.

Segment 2 adjusted losses.

\$240{,}000(1.25)(1.03) =\$300{,}000(1.03) =\$309{,}000.

Total adjusted losses are $ \$378{,}000+\$309{,}000=\$687{,}000 $. The on-level premiums are:

\begin{aligned} \$400{,}000(1.08)&=\$432{,}000,\\ \$300{,}000(1.02)&=\$306{,}000. \end{aligned}

Total on-level premium is $ \$432{,}000+\$306{,}000=\$738{,}000 $, so:

\frac{\$687{,}000}{\$738{,}000} =0.930894\ldots =93.09\%.

Check. Applying the 1.25 development factor to the first segment would incorrectly treat its already ultimate losses as undeveloped. The calculation uses each segment's stated loss basis and rounds the final percentage to two decimal places.

Common Mistakes and Exam Traps

The most important check is to ask, for each factor, what quantity it transforms. Development and trend act on losses in this calculation; on-leveling acts on premium. A factor placed on the wrong side can produce a plausible-looking percentage while answering a different question.

Omitting the premium adjustment. In the first worked example, multiplying the observed loss ratio by only the loss factors gives $ (480{,}000/900{,}000)(1.10)(1.04)=0.609067\ldots=60.91\% $. That omits the 1.05 on-level factor. The correct calculation divides by that factor, giving $58.11\%$.
Applying the on-level factor to losses. This changes the loss amount rather than restating the premium. Keep $O$ in the premium denominator: $LDT/(PO)$.
Applying development twice. If the selected losses are already ultimate, use a development factor of 1 for that basis, rather than applying a further factor.
Using an unweighted average for different groups. A factor average does not necessarily represent the aggregate adjustment. Add adjusted losses and on-level premiums separately, as in the group example.
Forgetting the units or rounding too early. Keep dollar amounts in the same units and retain precision through the division. Round only the final percentage to the requested precision.

Key takeaway: The combined adjusted loss ratio is total losses after the applicable development and trend adjustments divided by total premium after on-leveling. For heterogeneous groups, aggregate the adjusted amounts—not unadjusted totals multiplied by unweighted average factors.

Check Your Understanding

For each problem, determine which factors apply to losses and which apply to premium. Round percentage answers to two decimal places.

A hypothetical bicycle-coverage portfolio has $ \$200{,}000 $ of observed losses, a development factor of 1.15, a trend factor of 1.02, $ \$500{,}000 $ of earned premium, and an on-level factor of 1.04. Find its combined adjusted loss ratio. (Answer: 45.12%)
A hypothetical event-insurance segment has selected ultimate losses of $ \$360{,}000 $, a trend factor of 1.05, and $ \$600{,}000 $ of earned premium with an on-level factor of 1.08. Find the combined adjusted loss ratio. (Answer: 58.33%)
A hypothetical property portfolio has two groups. Group A has losses of $ \$180{,}000 $, development factor 1.10, trend factor 1.05, premium of $ \$300{,}000 $, and on-level factor 1.02. Group B has losses of $ \$220{,}000 $, development factor 1.05, trend factor 1.02, premium of $ \$400{,}000 $, and on-level factor 1.05. Find the aggregate adjusted loss ratio. (Answer: 61.09%)
A hypothetical equipment program has $ \$100{,}000 $ of observed losses with development factor 1.20 and trend factor 1.10, plus $ \$300{,}000 $ of observed losses with development factor 1.10 and trend factor 1.00. The corresponding earned premiums are $ \$200{,}000 $ with on-level factor 1.05 and $ \$500{,}000 $ with on-level factor 1.10. Find the combined adjusted loss ratio. (Answer: 60.79%)

What You'll Learn