Grossing Up Loss Costs for Premium Loads

Why Grossing Up Is Necessary

A prospective premium must fund more than expected claims. It may also need to cover fixed expenses per exposure and provisions for expenses, profit, or contingencies that are each stated as a percentage of premium. The last group creates a circular-looking calculation: the amount of each provision depends on the premium, but the premium must be large enough to pay those provisions as well as the underlying costs.

The solution is to write the premium balance explicitly and solve for premium. This is called grossing up. The key is to distinguish a dollar cost per exposure from a proportion of the gross premium. A percentage that is based on premium cannot simply be added to the expected loss cost as if it were a fixed dollar amount.

Definition: For one exposure on a consistent prospective basis, let $L$ be the expected loss cost, including loss adjustment expenses if they are part of the indicated cost; let $F$ be fixed expenses per exposure; and let $G$ be the gross premium per exposure. Let $r_1,\ldots,r_m$ be nonnegative provisions stated as fractions of gross premium. The total proportional load is $q=\sum_{j=1}^{m}r_j$. The balance equation is $G=L+F+qG$.

For example, an acquisition expense rate and a profit provision may both be expressed as percentages of gross premium. If they are 12% and 5%, respectively, then together they require $0.17G$ per exposure. By contrast, a fixed policy fee of $25 per exposure remains $25 in the balance equation; it is not multiplied by $G$.

Deriving the Gross Premium Formula

Start with the funding balance. Premium income per exposure must pay the expected loss cost, fixed expenses, and each premium-based provision. Since the $j$th provision is $r_jG$, the total of those provisions is $qG$.

G=L+F+\sum_{j=1}^{m}r_jG=L+F+qG

Move the premium-based amount to the left-hand side and factor out $G$. When $q<1$, division gives the unique gross premium:

G(1-q)=L+F \qquad\Longrightarrow\qquad G=\frac{L+F}{1-q} =\frac{L+F}{1-\sum_{j=1}^{m}r_j}.

This derivation explains the denominator adjustment: each additional dollar of gross premium also generates $q$ dollars of proportional provisions, so only $1-q$ dollars of each premium dollar remain to fund $L+F$. The result applies when the rates are all defined on the same gross-premium base. If a provision is based on losses, earned premium, or another quantity instead, it cannot be included in $q$ without first expressing it consistently.

Theorem: If $L+F>0$, all proportional rates are nonnegative, and $q<1$, the balance equation has the finite positive solution $G=(L+F)/(1-q)$. If $q=1$, no finite premium satisfies the balance when $L+F>0$. If $q>1$, the algebraic solution is negative, so there is no economically meaningful positive premium under this model.

The condition $q<1$ is more than a technical detail. It says that some fraction of each premium dollar must remain after proportional provisions to fund expected losses and fixed expenses. A total proportional load close to 100% is mathematically possible when it is still below 100%, but it produces a very large gross-up. The formula also assumes a common exposure basis: $L$, $F$, and $G$ must all refer to the same exposure unit and prospective period.

Worked Applications

Worked Example: Grossing Up a Discrete Expected Loss Cost

A hypothetical annual policy has aggregate covered loss $X$ with distribution $P(X=0)=0.75$, $P(X=\$400)=0.20$, and $P(X=\$1{,}800)=0.05$. Expected loss adjustment expense is $30 per exposure, and fixed expense is $40 per exposure. Acquisition expense is 22% of gross premium, other variable expense is 8%, and the profit provision is 5%. Find the gross premium per exposure. Assume no investment income adjustment and no separate mortality model; all amounts and rates are on the same annual exposure basis.

Method. Calculate the expected loss cost, add the fixed per-exposure amounts, and divide by the fraction of premium left after proportional provisions.

First calculate the expected covered loss:

E[X]=0(0.75)+400(0.20)+1{,}800(0.05) =0+80+90=\$170.

Add expected loss adjustment expense to obtain $L=\$170+\$30=\$200$. Add fixed expense to obtain the non-premium-based amount $L+F=\$200+\$40=\$240$. The proportional rates sum to $q=0.22+0.08+0.05=0.35$, so the fraction available to fund that $240 is $1-q=0.65$.

G=\frac{240}{0.65}=\$369.230769\ldots

To check, the proportional provisions total $0.35(\$369.230769\ldots)=\$129.230769\ldots$. Adding the non-premium costs gives $\$240+\$129.230769\ldots=\$369.230769\ldots$, which equals the gross premium. Rounded only at the final step, the gross premium is $369.23 per exposure.

Common mistake avoided: Adding 35% of the $240 cost base would give only $324. That calculation applies premium-based rates to the wrong base. The provisions are percentages of $G$, not percentages of $L+F$.

Worked Example: Several Premium-Based Provisions

For a hypothetical annual exposure, expected losses and loss adjustment expenses together are $510. Fixed expenses are $55. The expense provisions are 17% for acquisition, 6% for administration, and 3.5% for taxes and assessments; a contingency and profit provision is 3%. Each stated rate is a percentage of gross premium. Find the required gross premium. No interest or mortality adjustment is included.

Method. Combine all provisions that share the gross-premium base, then apply the denominator formula. The fixed amount stays outside the percentage sum.

The non-premium-based amount is $\$510+\$55=\$565$. The total proportional rate is:

$$ q=0.17+0.06+0.035+0.03=0.295. $$

Thus $1-q=1-0.295=0.705$, which is positive, and:

G=\frac{565}{0.705}=\$801.418439\ldots

The total premium-based provisions are $0.295(\$801.418439\ldots)=\$236.418439\ldots$. The check is $\$565+\$236.418439\ldots=\$801.418439\ldots$. Therefore the gross premium is $801.42 per exposure, rounded to the nearest cent.

Common mistake avoided: Adding the rates correctly is essential: 17% plus 6% plus 3.5% plus 3% is 29.5%, not 26.5%. Omitting the contingency and profit provision would understate the required gross premium.

Worked Example: A High Load That Still Permits a Finite Premium

A hypothetical annual portfolio has aggregate loss $X=\$0$ with probability 0.40 and $X=\$100$ with probability 0.60. Fixed expense is $40 per exposure. The premium-based expense provision is 68% of gross premium, and the profit provision is 22% of gross premium. Find the gross premium per exposure. Assume no loss adjustment expense, no investment income adjustment, and no mortality model.

Method. Calculate the expected loss, form the non-premium cost, and check the total proportional rate before dividing.

The expected loss is $E[X]=0(0.40)+100(0.60)=\$60$. Therefore, $L+F=\$60+\$40=\$100$. The proportional rate total is $q=0.68+0.22=0.90$, leaving $1-q=0.10$ of premium to fund the $100 cost.

G=\frac{100}{1-0.90}=\frac{100}{0.10}=\$1{,}000.

Check the balance: the expense provision is $0.68(\$1{,}000)=\$680$, and the profit provision is $0.22(\$1{,}000)=\$220$. The total is $\$100+\$680+\$220=\$1{,}000$, as required. The gross premium is $1,000 per exposure.

Common mistake avoided: It is wrong to conclude that no finite premium exists merely because the provisions take 90% of premium. Since 90% is less than 100%, the remaining 10% can fund the $100 cost. The premium is large, but finite.

Interpreting the Result and Avoiding Base Errors

Grossing up is especially sensitive to the definition of each rate. A percentage stated as “of premium” must be applied to gross premium under the balance equation above. A percentage stated as “of losses” has a different base and is not automatically one of the $r_j$ terms. Likewise, a fixed dollar expense should not be turned into a percentage merely to make the formula look uniform.

A useful exam check is to substitute the calculated premium back into the original balance. Calculate each provision using its stated base, add the dollar costs, and confirm that the total equals $G$. This catches both an incorrect denominator and a misread expense basis.

Common exam trap: If $q=0.90$, the denominator is $1-q=0.10$, not $0.90$. For $100 of costs, the wrong calculation $100/0.90=\$111.11$ leaves only $11.11 after a 90% provision, so it cannot fund the $100 cost. The corrected calculation is $100/(1-0.90)=\$1{,}000$, and 10% of that premium is exactly $100.

Another diagnostic is to compare the calculated premium with the costs it must fund. With positive proportional rates and positive $L+F$, the gross premium must exceed $L+F$. If a calculation produces a smaller premium, or if the balance check fails, revisit the rate sum and the bases. If the sum of premium-based provisions reaches or exceeds 100%, do not interpret a negative or undefined algebraic result as a usable premium; under these assumptions, a positive finite gross premium cannot fund the stated costs.

The method is aligned with the SOA Exam ASTAM (Advanced Short-Term Actuarial Mathematics) syllabus. Its central discipline is to build the funding equation from the stated cost bases before doing arithmetic: fixed amounts enter directly, while percentages of gross premium are collected into the denominator adjustment.

Check Your Understanding

For each hypothetical annual exposure, find the gross premium or determine whether a positive finite gross premium exists. Assume every stated percentage is based on gross premium unless the question says otherwise. Round monetary answers to the nearest cent.

Aggregate loss $X$ is $0 with probability 0.70, $300 with probability 0.20, and $1,000 with probability 0.10. Fixed expense is $20, and the only premium-based provisions are 12% variable expense and 8% profit. (Answer: $225.00 per exposure)
Expected losses and loss adjustment expenses are $375, fixed expense is $45, and the provisions are 14% acquisition expense, 9% other variable expense, and 2% profit. Find the gross premium. (Answer: $560.00 per exposure)
Aggregate loss $X$ is $0 with probability 0.60, $250 with probability 0.30, and $1,000 with probability 0.10. Fixed expense is $25, with premium-based provisions totaling 40%. Find the gross premium. (Answer: $333.33 per exposure)
Expected losses and loss adjustment expenses are $420, and fixed expenses are $80. Premium-based expense provisions are 30% and 25%, with a 20% profit provision. Find the gross premium and state whether the denominator condition is satisfied. (Answer: $2,000.00 per exposure; yes, the rates total 75%, which is less than 100%)
Expected losses and loss adjustment expenses are $500, fixed expenses are $100, and the premium-based provisions are 65% and 35%. Does a finite positive gross premium exist under the stated model? (Answer: No; the rates total 100%, so the balance equation has no finite solution for positive costs)

What You'll Learn