Applying ILFs To Base-Limit Premiums

From a Base-Limit Loss Indication to a Higher-Limit Premium

An increased limits factor (ILF) converts an expected loss at one policy limit to an expected loss at another. The conversion applies to the loss component of the indication. A quoted or indicated premium may also contain fixed expenses and expenses that vary with premium, so multiplying the entire premium by an ILF generally changes components that the factor was not designed to change.

The practical task is therefore two-part: first adjust the base-limit expected loss using the appropriate ILF; then calculate the premium using the applicable expense provisions. This distinction is aligned with the CAS Exam 8 (Advanced Ratemaking) syllabus.

Definition: For a specified risk population and consistent coverage terms, the ILF for limit $u$ relative to base limit $b$ is the ratio of expected limited losses at those limits: $$ \operatorname{ILF}(u;b)=\frac{E[Y_u]}{E[Y_b]}. $$ Here, $Y_u$ and $Y_b$ denote the payment per loss under the respective limits. Thus, if $L_b$ is the expected loss at the base limit, the corresponding expected loss at limit $u$ is $L_u=L_b\operatorname{ILF}(u;b)$. The factor applies to expected losses on the same exposure and loss basis; it does not, by itself, adjust expenses, commissions, profit provisions, or other premium components.

This definition presumes the ILF and base-limit loss indication refer to compatible populations, time periods, and policy terms. If the ILF reflects a different deductible, coverage, or loss basis, the multiplication may not be appropriate without reconciling that difference. When the base-limit expected loss is split by class, territory, or another rating segment, apply the corresponding factor to each segment’s expected loss before summing.

Separate the Loss Calculation from the Premium Calculation

A useful notation is $L_b$ for base-limit expected loss, $F$ for fixed expenses, $v$ for variable expenses as a proportion of premium, and $P$ for premium. Under the simple expense model $P=L+F+vP$, the premium is $P=(L+F)/(1-v)$. The variable expense is a percentage of premium, not a percentage of expected loss.

L_u=L_b\operatorname{ILF}(u;b), \qquad P_u=\frac{L_u+F}{1-v}

These equations are a pricing structure, not a universal expense formula. Use them only when fixed expenses are stated as a dollar amount and the applicable variable expense provision is a fraction of premium, with $v<1$. Other expense or profit provisions may require a different premium equation. In every case, identify which components the ILF is intended to scale before applying it.

Establish the base loss.
Identify the expected loss at the base limit for the relevant exposure, class, and time period. Do not substitute total premium if it includes expenses.

Apply the ILF.
Multiply each base-limit expected loss by its matching factor. Sum the resulting higher-limit expected losses if there are multiple segments.

Complete the premium indication.
Add fixed expenses and incorporate variable expenses and any other stated provisions using their specified bases.

Why the ILF Scales Expected Loss

For a fixed group of risks, let the expected loss per exposure at the base limit be $E[Y_b]$, and let the expected loss per exposure at the higher limit be $E[Y_u]$. By the ILF definition, $E[Y_u]=E[Y_b]\operatorname{ILF}(u;b)$. For $n$ comparable exposures, linearity of expectation gives the total expected loss at the higher limit as $nE[Y_u]=nE[Y_b]\operatorname{ILF}(u;b)$. No independence assumption is needed for this expectation identity; the population and loss basis must simply be consistent.

The premium does not generally obey the same scaling. For example, under the expense model above, $P_b=(L_b+F)/(1-v)$ and $P_u=(\operatorname{ILF}\cdot L_b+F)/(1-v)$. The fixed expense is present in both premiums but is not multiplied by the ILF. Therefore, when $F>0$, the premium ratio $P_u/P_b$ will generally differ from the ILF.

Worked Example: One Class with Fixed and Variable Expenses

A hypothetical portfolio has 125 policies. Expected loss at the base limit is $480 per policy, and the ILF to the requested higher limit is 1.35. Fixed expense is $36 per policy; variable expense is 7% of premium. Find the total higher-limit premium and compare the base- and higher-limit premium ratio with the ILF.

Method and assumptions. Treat the stated expected loss as the base-limit loss component. Assume the ILF applies to that loss, fixed expense stays at $36 per policy, and the 7% variable expense is a percentage of premium.

Base-limit amounts. Base expected loss is $125\times \$480=\$60{,}000$. Fixed expense is $125\times \$36=\$4{,}500$. With $v=0.07$, the base-limit premium indication is $(\$60{,}000+\$4{,}500)/(1-0.07)=\$64{,}500/0.93=\$69{,}354.84$.

Higher-limit loss. The higher-limit expected loss is $\$60{,}000\times1.35=\$81{,}000$. The fixed expense remains $4,500. Thus $P_u=(\$81{,}000+\$4{,}500)/0.93=\$85{,}500/0.93=\$91{,}935.48$. The variable expense included in this premium is $0.07\times \$91{,}935.48=\$6{,}435.48$, so loss, fixed expense, and variable expense sum to $91,935.48.

Check and trap avoided. The ratio of the premiums is $\$91{,}935.48/\$69{,}354.84=1.325$, not 1.35. Multiplying the entire base premium by 1.35 would also scale the fixed expense and would not preserve the stated expense structure. The factor correctly scales expected loss only.

Portfolio Segments Require Loss-Weighted Aggregation

When a portfolio contains segments with different base expected losses and different ILFs, calculate higher-limit loss by segment. An overall ILF can be derived afterward as the ratio of total higher-limit loss to total base-limit loss. This is a base-loss-weighted factor, not generally the simple arithmetic average of the segment factors.

\operatorname{ILF}_{\text{portfolio}} = \frac{\sum_j L_{b,j}\operatorname{ILF}_j}{\sum_j L_{b,j}}

The weights are the segments’ base expected losses. A factor from a segment with a large base expected loss contributes more to the portfolio result than the same-sized factor from a segment with a small base expected loss.

Worked Example: Combining Two Classes

A hypothetical insurer has 40 policies in Class A, each with base-limit expected loss of $250 and an ILF of 1.20. Class B has 60 policies, each with base-limit expected loss of $400 and an ILF of 1.05. Fixed expense is $20 per policy, and variable expense is 4% of premium. Find total higher-limit expected loss and total premium, and calculate the portfolio ILF.

Class A. Base expected loss is $40\times \$250=\$10{,}000$. Higher-limit expected loss is $\$10{,}000\times1.20=\$12{,}000$.

Class B. Base expected loss is $60\times \$400=\$24{,}000$. Higher-limit expected loss is $\$24{,}000\times1.05=\$25{,}200$.

Portfolio loss and factor. Total base expected loss is $\$10{,}000+\$24{,}000=\$34{,}000$. Total higher-limit expected loss is $\$12{,}000+\$25{,}200=\$37{,}200$. Therefore, the portfolio ILF is $\$37{,}200/\$34{,}000=1.0941$ to four decimal places. The simple average, $(1.20+1.05)/2=1.125$, is not the portfolio ILF because the two class base losses are unequal.

Premium. Fixed expense is $100\times \$20=\$2{,}000$. With variable expense ratio $v=0.04$, total premium is $(\$37{,}200+\$2{,}000)/(1-0.04)=\$39{,}200/0.96=\$40{,}833.33$.

Check and trap avoided. The variable expense is $0.04\times \$40{,}833.33=\$1{,}633.33$; adding it to $37,200 of expected loss and $2,000 of fixed expense gives $40,833.33. Averaging the class factors without weighting, or applying one unweighted factor to total base loss, would misstate the loss indication.

When the Information Is Given as a Base Premium

Sometimes the available figure is a base-limit premium rather than a base-limit expected loss. A total premium alone is not enough to apply an ILF correctly: it does not reveal how much represents expected loss and how much represents expenses or other provisions. If the expense formula and its components are supplied, recover the loss component first. Then apply the ILF and rebuild the premium at the higher limit.

Worked Example: Extracting Loss from a Base Premium

A hypothetical group of 80 policies has an indicated base-limit premium of $42,000. Fixed expense is $25 per policy, and variable expense is 8% of premium. The ILF to a higher limit is 1.30. Assuming the base premium follows $P_b=(L_b+F)/(1-v)$, calculate the higher-limit premium.

Method and assumptions. The $42,000 is a total premium that includes the specified fixed and variable expenses. No other premium components are included. Recover the aggregate base expected loss before applying the factor.

Recover the base loss. Total fixed expense is $80\times \$25=\$2{,}000$. From $P_b=(L_b+F)/(1-v)$, rearrange to $L_b=(1-v)P_b-F$. Thus $L_b=0.92\times \$42{,}000-\$2{,}000=\$38{,}640-\$2{,}000=\$36{,}640$.

Apply the ILF. Higher-limit expected loss is $\$36{,}640\times1.30=\$47{,}632$. Fixed expense remains $2,000, so $P_u=(\$47{,}632+\$2{,}000)/0.92=\$49{,}632/0.92=\$53{,}947.83$, rounded to the nearest cent.

Check and trap avoided. The variable expense at the higher limit is $0.08\times \$53{,}947.83=\$4{,}315.83$. Adding $47,632, $2,000, and $4,315.83 gives $53,947.83. Multiplying $42,000 directly by 1.30 would incorrectly scale the expenses embedded in the base premium.

Common Premium-Calculation Trap

The following incorrect calculation illustrates why the distinction between loss and premium matters. Suppose there are 80 policies, base expected loss is $500 per policy, the ILF is 1.30, fixed expense is $25 per policy, and variable expense is 8% of premium. A proposed method first adds fixed expense to base loss, then multiplies that amount by the ILF.

Wrong calculation: Base loss plus fixed expense is $80\times(\$500+\$25)=\$42{,}000$. Multiplying this by 1.30 and grossing up gives $(\$42{,}000\times1.30)/0.92=\$59{,}347.83$. The error is that the ILF has been applied to fixed expenses as if they were expected losses.

Correction: Higher-limit expected loss is $80\times \$500\times1.30=\$52{,}000$. Total fixed expense is $80\times \$25=\$2{,}000$. Therefore, the correct premium is $(\$52{,}000+\$2{,}000)/0.92=\$58{,}695.65$. The variable expense is $0.08\times \$58{,}695.65=\$4{,}695.65$; together, $52,000, $2,000, and $4,695.65 verify the premium to the nearest cent.

The key diagnostic is to label every amount before calculating: base expected loss, higher-limit expected loss, fixed expense, and variable expense. If the input is a total premium, identify or recover its loss component before applying an ILF. If the relevant component breakdown is unavailable, the higher-limit premium cannot be determined from the total premium and ILF alone.

Key takeaway: Apply an ILF to the expected-loss component on a consistent basis. Aggregate segment losses using their respective factors, then apply the stated expense structure. A higher limit changes expected loss; it does not automatically multiply fixed expenses or every other component of premium.

Check Your Understanding

For each hypothetical indication, separate expected loss from expenses before finding the requested result. Round dollar premiums to the nearest cent.

A portfolio has 50 policies with base expected loss of $320 per policy. The ILF is 1.15, fixed expense is $18 per policy, and variable expense is 5% of premium. Find the total higher-limit premium. (Answer: $20,315.79)
Class A has 25 policies with base expected loss of $800 each and an ILF of 1.20. Class B has 75 policies with base expected loss of $300 each and an ILF of 1.08. Fixed expense is $15 per policy and variable expense is 6% of premium. Find total higher-limit expected loss and total premium. (Answer: expected loss $48,300; premium $52,978.72)
A group has 120 policies with base expected loss of $400 per policy, an ILF of 1.25, fixed expense of $10 per policy, and variable expense of 5% of premium. Find the higher-limit premium and explain why multiplying a base premium by 1.25 would not generally be valid. (Answer: $64,421.05; the ILF scales expected loss, not fixed expense or the full premium)
A portfolio has 60 policies, base expected loss of $600 per policy, a uniform loss trend factor of 1.05, and an ILF of 1.25. Fixed expense is $30 per policy and variable expense is 7% of premium. Assume the trend applies to the base expected-loss estimate before the ILF, with no other adjustments. Find the higher-limit premium. (Answer: $52,741.94)
A group of 100 policies has a base-limit total premium of $50,000. Fixed expense is $15 per policy, variable expense is 10% of premium, and the ILF is 1.20. Assuming the stated expense model includes all premium components, recover the base expected loss and find the higher-limit premium. (Answer: base expected loss $43,500; higher-limit premium $59,666.67)

What You'll Learn