Why the Premium Basis Matters
A territorial loss ratio compares losses with the premium associated with the exposures that produced those losses. Its interpretation depends on both sides of the ratio: a territory can have a different loss ratio because its losses differ, because its premium differs, or because both differ. In particular, premium earned under historical rates is not automatically comparable across territories or periods when the applicable rates have changed.
For a ratemaking comparison, premium is commonly adjusted to a consistent rate level before calculating the ratio. This isolates the comparison from changes in the amount of premium produced solely by rate changes. It does not, by itself, remove differences in exposure mix or show which territory has the lower expected loss per exposure.
Adjusting Earned Premium to a Consistent Rate Level
Suppose a block of policies earned premium while an earlier rate level was in effect. If the applicable current rates are higher, the premium is increased to estimate what that same business would have produced at the current level, holding its exposures and other rating characteristics constant. If rates decreased, the corresponding adjustment factor is less than 1.
For premium earned under rate level \(j\), define \(f_j\) as the factor that converts that premium to the applicable current level. In the simplest case, \(f_j\) is the current average rate divided by the average rate in effect for that premium segment. Then the premium at current rate level is the sum of each segment's earned premium multiplied by its own factor.
Here \(P_j\) is earned premium for segment \(j\), and \(L\) is the losses for the matching experience. When rate changes differ by territory, the applicable factors must reflect the rates relevant to each territory; a single statewide factor is not automatically appropriate. If the premium has already been restated to the current level, do not adjust it a second time.
The calculation assumes that the factor represents the rate change for the premium being adjusted and that exposures, coverage, and other rating characteristics are held constant in estimating premium at the new level. If the mix of business or coverage has changed, a simple rate-level factor does not correct that change. Premium and losses must also be matched by experience period: for example, do not divide losses from earned exposures by an unrelated written-premium total.
Why the Adjustment Works
The reason for multiplying historical premium by a rate-level factor follows directly from the meaning of a rate change. For a fixed exposure with unchanged coverage and rating characteristics, premium is proportional to the rate. If the rate rises by a factor \(f\), the premium for that same exposure also rises by \(f\). Applying that relationship to each segment gives the adjusted premium.
The losses in the numerator are not multiplied by this premium rate-level factor. The adjustment asks what premium the observed exposures would have generated at the comparison rate level; it does not change the losses that occurred. Consequently, increasing the premium denominator lowers the calculated loss ratio, all else equal.
Worked Example: Combining Premium at Several Rate Levels
A hypothetical territory has two earned-premium segments. It earned $500,000 at an older rate level indexed at 1.00 and $300,000 at an intermediate level indexed at 1.10. The applicable current level is 1.21. Losses for the matching experience are $748,000. Calculate the current-level premium and loss ratio. Then compare the result with the unadjusted ratio.
Method. Adjust each premium segment separately, using the ratio of current to historical rate level. The older segment's factor is \(1.21/1.00=1.21\). The intermediate segment's factor is \(1.21/1.10=1.10\).
The older segment becomes \( \$500{,}000 \times 1.21 = \$605{,}000\). The intermediate segment becomes \( \$300{,}000 \times 1.10 = \$330{,}000\). Therefore the current-level premium is \( \$605{,}000+\$330{,}000=\$935{,}000\).
The current-level loss ratio is \( \$748{,}000/\$935{,}000=0.8000\), or 80.00%. The unadjusted earned premium is \( \$500{,}000+\$300{,}000=\$800{,}000\), giving an unadjusted ratio of \( \$748{,}000/\$800{,}000=0.9350\), or 93.50%.
Check. The current-level premium is greater than the historical premium because both rate-level factors exceed 1. The adjusted loss ratio is therefore lower, as expected. The common mistake avoided here is treating the two segments as if they shared one historical rate level.
Premium Differences Can Change Territory Comparisons
A loss ratio is a measure of losses relative to premium, not a direct measure of losses per exposure. Consider two territories with comparable coverage and current-level premium, but different premiums per exposure. The territory with the higher premium can show a lower loss ratio even if it has higher losses per exposure. That lower ratio may reflect the larger premium denominator rather than lower underlying losses.
This distinction is important when interpreting a comparison. A loss ratio can indicate how losses compare with the premium base used in the calculation. It does not establish that one territory is intrinsically safer or has lower expected losses per exposure. For that kind of question, the exposure basis and the comparability of the underlying business must also be considered.
Worked Example: A Lower Loss Ratio Does Not Mean Lower Losses per Exposure
Two hypothetical territories have current-level earned premium and losses for the same coverage. Territory A has 1,000 exposures, $600,000 of losses, and $750,000 of premium. Territory B has 800 exposures, $520,000 of losses, and $800,000 of premium. Calculate each territory's loss ratio and losses per exposure. Explain why the rankings differ.
Territory A. Its loss ratio is \( \$600{,}000/\$750{,}000=0.8000\), or 80.00%. Its losses per exposure are \( \$600{,}000/1{,}000=\$600\).
Territory B. Its loss ratio is \( \$520{,}000/\$800{,}000=0.6500\), or 65.00%. Its losses per exposure are \( \$520{,}000/800=\$650\).
By loss ratio, B is lower. By losses per exposure, A is lower. The premium per exposure is \( \$750{,}000/1{,}000=\$750\) in A and \( \$800{,}000/800=\$1{,}000\) in B. B's larger premium per exposure helps produce its lower ratio, even though its losses per exposure are higher.
Check. Multiplying each loss ratio by its premium per exposure recovers its losses per exposure: \(0.80\times\$750=\$600\) for A and \(0.65\times\$1{,}000=\$650\) for B. The common mistake avoided is interpreting a lower loss ratio as proof of lower losses per exposure.
Use the Applicable Factor for Each Territory
A rate-level factor must correspond to the rates applicable to the premium being adjusted. If territories had different rate changes, applying one territory's factor to all territories can distort the comparison. The following example shows why each territory's earned premium should be adjusted on its own applicable basis.
Worked Example: Comparing Territories With Different Rate Changes
Two hypothetical territories have earned premium from different historical rate levels. Territory East earned $600,000 when the applicable factor to the current level is 1.10, and its matching losses are $528,000. Territory West earned $500,000 when its applicable factor is 1.20, and its matching losses are $480,000. Calculate both current-level loss ratios and compare them with the unadjusted ratios.
East. Current-level premium is \( \$600{,}000\times1.10=\$660{,}000\). Its current-level ratio is \( \$528{,}000/\$660{,}000=0.8000\), or 80.00%. Its unadjusted ratio is \( \$528{,}000/\$600{,}000=0.8800\), or 88.00%.
West. Current-level premium is \( \$500{,}000\times1.20=\$600{,}000\). Its current-level ratio is \( \$480{,}000/\$600{,}000=0.8000\), or 80.00%. Its unadjusted ratio is \( \$480{,}000/\$500{,}000=0.9600\), or 96.00%.
On the unadjusted basis, West appears to have the higher loss ratio. At the applicable current level, the ratios are equal. The difference in the unadjusted comparison was affected by the different rate-level changes, not just by the losses.
Check. Both adjusted denominators are larger than their historical premiums, and both adjusted ratios are lower than their unadjusted ratios. Applying East's factor to West would give West a premium of $550,000 and a ratio of 87.27%, an incorrect result because 1.10 is not West's applicable factor.
Common Mistakes and Interpretation Limits
A useful calculation can still be misinterpreted if the denominator or the comparison question is not stated clearly. Before comparing results, verify the premium basis and be explicit about what the ratio can—and cannot—say.
- Dividing premium by the factor. If rates increased by a factor of 1.20, current-level premium is historical premium multiplied by 1.20, not divided by 1.20. For example, $100,000 becomes $120,000, not $83,333.
- Using one factor for unlike segments. Apply each premium segment's relevant factor before summing. A factor based on a different territory or rate change does not match that segment.
- Adjusting twice. If the earned premium has already been expressed at the applicable current level, another rate-level adjustment overstates the denominator.
- Mixing premium bases. Match the losses to the earned premium for the associated exposures and period. A written-premium total is not interchangeable with earned premium without an appropriate basis for doing so.
- Reading the ratio as a risk measure alone. A lower loss ratio can result from higher premium rather than lower losses per exposure. A loss ratio alone does not isolate exposure mix or establish a territory's underlying risk level.
Check Your Understanding
For each problem, calculate the requested ratio using the specified premium basis. Monetary amounts are hypothetical.
- A territory earned $400,000 at a historical rate level. The factor to the applicable current level is 1.15, and matching losses are $368,000. What is its current-level loss ratio? (Answer: $460,000 current-level premium; 80.00% loss ratio.)
- Territory North has $525,000 of losses, $700,000 of earned premium, and a current-level factor of 1.08. Territory South has $600,000 of losses and $750,000 of earned premium already at the current level. Calculate both current-level loss ratios and identify the lower one. (Answer: North 69.44%; South 80.00%; North is lower.)
- Two territories each have $420,000 of matching losses. Territory C has $500,000 of earned premium with a factor of 1.20; Territory D has $600,000 of premium already at the current level. Calculate both current-level ratios and the unadjusted ratios. (Answer: Both current-level ratios are 70.00%; unadjusted, C is 84.00% and D is 70.00%.)
- A territory has $720,000 of losses. Its earned premium consists of $300,000 with factor 1.20 and $400,000 with factor 1.05. Calculate the current-level premium and loss ratio. (Answer: $780,000 premium; 92.31% loss ratio.)
- Territory A has $450,000 of losses and $500,000 of earned premium with a current-level factor of 1.10. Territory B has $490,000 of losses and $560,000 of premium already at the current level. Compare both the unadjusted and current-level loss ratios. (Answer: Unadjusted, A is 90.00% and B is 87.50%; current-level, A is 81.82% and B is 87.50%, so the ranking reverses.)