Technical Price Components for Reinsurance

Why a Reinsurance Quote Needs a Component View

A reinsurance premium is not simply an estimate of the recoveries the reinsurer expects to pay. A quote may also provide for uncertainty in those recoveries, the reinsurer’s expenses, brokerage, and a target profit. Separating these components makes the calculation auditable: it shows which assumptions drive the indicated price and helps explain why a market quote may differ from that indication.

This tutorial focuses on the price decomposition, not on estimating treaty recoveries from a layer or a full loss model. Assume the expected recovery and, where needed, its uncertainty have already been estimated for the contract being priced. The methods are aligned with the CAS Exam 9 (Risk Management for Actuaries) syllabus.

Definition: A technical price is a premium indication calculated from an explicit set of assumptions about expected recoveries, uncertainty or risk provision, expenses, brokerage, and target profit. A market price is the premium actually quoted or agreed in the market. The two can differ because of differences in assumptions, competition, capacity, negotiation, or other commercial considerations.

Core Components and a Consistent Pricing Basis

Let $R$ denote the reinsurer’s recovery payment over the pricing period. The expected recovery, $E[R]$, is the modeled average payment under the assumed loss distribution. It is the expected claims cost, not a guarantee of what the reinsurer will pay in a particular year.

A risk load is an additional provision for bearing uncertainty in the recovery. Its amount depends on the chosen pricing method and assumptions. For example, a standard-deviation loading uses a nonnegative factor $k$ multiplied by the standard deviation of $R$. This is one possible pricing convention, not a universal rule. It should not be added if the same uncertainty has already been allowed for elsewhere.

Expenses can be represented as a fixed dollar amount or as a percentage of premium. Brokerage is commonly modeled as a percentage of the gross premium. A target profit provision can also be specified as a percentage of gross premium. The pricing equation must state the basis for each item: applying a percentage to gross premium is different from adding a fixed dollar amount.

Component	Typical basis in a simplified model	Pricing interpretation
Expected recovery	Expected dollars paid	Modeled average claims cost
Risk load	Dollars, often calculated from a risk measure	Provision for uncertainty beyond the mean
Fixed expenses	Dollars	Costs not modeled as a share of premium
Brokerage	Percentage of gross premium	Commission included in the price
Target profit	Dollars or percentage of gross premium	Intended return after the other specified provisions

Suppose the expected recovery is $E[R]$, the risk load is $L$, and fixed expenses are $e$. Let $b$ be the brokerage rate and $p$ the target profit rate, both applied to gross premium $G$. In this model, the premium must cover the dollar costs and the percentage-of-premium provisions.

$$ G=E[R]+L+e+bG+pG $$

Rearranging gives the gross-up formula. It applies only when the percentage loads are defined on gross premium and $b+p<1$; it does not apply unchanged if a percentage is charged on a different base or if expenses are modeled differently.

G(1-b-p)=E[R]+L+e \qquad\Longrightarrow\qquad G=\frac{E[R]+L+e}{1-b-p}

The equation explains why adding the dollar costs and then stopping is generally incorrect. If brokerage and profit are percentages of $G$, they depend on the final premium, so the premium must be solved for rather than treated as a known amount.

How an Uncertainty Load Enters the Indication

A model may use a risk measure of the recovery distribution to set a risk load. For illustration, the standard-deviation principle sets $L=k\sqrt{\operatorname{Var}(R)}$, where $k$ is a nonnegative pricing factor chosen for the application. The formula states a pricing convention; it does not claim that every reinsurer uses this method or that the factor can be selected without judgment.

L=k\sqrt{\operatorname{Var}(R)}

The rationale is that a distribution with greater spread around the same mean can expose the reinsurer to greater variability in payments. The standard deviation expresses that spread in the same units as the recovery, so multiplying it by a dimensionless factor gives a dollar load. The factor is a pricing assumption, not a probability or a margin that can be inferred from the recovery distribution alone.

Process variability in future losses and uncertainty about the model’s parameters are distinct concerns. A chosen risk load may be intended to reflect one or both, but the pricing basis should say what it represents. Avoid charging twice for the same source of uncertainty—for example, by adding a separate contingency provision after a risk load that already includes it.

Set the modeled recovery basis.
State what contract and pricing period $R$ represents, and identify the distribution or estimates used to obtain $E[R]$.

Choose and calculate the risk provision.
Specify the risk measure and loading assumption. Check that the load is not duplicating another provision.

Classify each non-claims component.
Record whether expenses are fixed dollars or percentages, and state the base for brokerage and target profit.

Solve and reconcile.
Calculate the gross indication, then compare it with any market quote using the same component definitions and bases.

Worked Price Decompositions

Worked Example: Grossing Up a Discrete Recovery Model

A hypothetical treaty has recovery $R$, measured in thousands of dollars, with distribution $P(R=0)=0.50$, $P(R=1{,}000)=0.30$, and $P(R=3{,}000)=0.20$. Use a standard-deviation load with $k=0.15$, fixed expenses of $40$ thousand dollars, brokerage of $4\%$ of gross premium, and target profit of $6\%$ of gross premium. Calculate the technical gross premium in thousands of dollars, rounded to two decimals. The method is to calculate the recovery moments first and then gross up the dollar provisions. This avoids treating the risk load as a percentage of expected recovery.

Expected recovery. The probability-weighted mean is $E[R]=(0.50)(0)+(0.30)(1{,}000)+(0.20)(3{,}000)=0+300+600=900$.

Recovery variance. The second moment is $E[R^2]=(0.50)(0^2)+(0.30)(1{,}000^2)+(0.20)(3{,}000^2)=0+300{,}000+1{,}800{,}000=2{,}100{,}000$. Thus, $\operatorname{Var}(R)=2{,}100{,}000-(900)^2=2{,}100{,}000-810{,}000=1{,}290{,}000$, in squared thousands of dollars.

Risk load. The standard deviation is $\sqrt{1{,}290{,}000}=1{,}135.78$ thousand dollars. Therefore $L=(0.15)(1{,}135.78)=170.37$ thousand dollars.

Gross premium. The dollar provisions before percentage loads total $900+170.37+40=1{,}110.37$ thousand dollars. The percentage loads sum to $0.04+0.06=0.10$, so $G=1{,}110.37/(1-0.10)=1{,}233.74$ thousand dollars.

The indication is $1{,}233.74$ thousand dollars, or $\$1{,}233{,}740$ at the stated precision. As a check, brokerage is $0.04(1{,}233.74)=49.35$, and target profit is $0.06(1{,}233.74)=74.02$, in thousands. Adding $900+170.37+40+49.35+74.02=1{,}233.74$, subject to rounding. The common mistake this method avoids is adding $10\%$ to the dollar provisions once; that would not account for percentage loads being calculated on the final premium.

Worked Example: Including a Percentage Expense

For a hypothetical cover, recovery $R$, in thousands of dollars, has distribution $P(R=0)=0.70$, $P(R=400)=0.20$, and $P(R=1{,}200)=0.10$. Use $k=0.25$ for a standard-deviation load, fixed expenses of $12$ thousand dollars, variable expenses of $3\%$ of gross premium, brokerage of $7\%$, and target profit of $5\%$, all percentage rates being on gross premium. Find the indication rounded to two decimals in thousands of dollars. This method separates fixed dollar costs from all percentage loads. It avoids mistakenly including the percentage expense only in the numerator.

Expected recovery. $E[R]=(0.70)(0)+(0.20)(400)+(0.10)(1{,}200)=0+80+120=200$ thousand dollars.

Risk load. $E[R^2]=(0.20)(400^2)+(0.10)(1{,}200^2)=32{,}000+144{,}000=176{,}000$. Thus $\operatorname{Var}(R)=176{,}000-(200)^2=136{,}000$, and the standard deviation is $\sqrt{136{,}000}=368.78$ thousand dollars. The load is $L=(0.25)(368.78)=92.20$ thousand dollars.

Gross premium. The fixed-dollar provisions total $200+92.20+12=304.20$. The variable expense, brokerage, and profit rates total $0.03+0.07+0.05=0.15$. Therefore $G=304.20/(1-0.15)=357.88$ thousand dollars.

The components check at the unrounded premium: variable expense is about $10.74$, brokerage is about $25.05$, and profit is about $17.89$, all in thousands. The total is approximately $304.20+10.74+25.05+17.89=357.88$. The common mistake is to use $304.20$ as the quote, even though that amount leaves out each percentage-of-premium component.

Worked Example: Comparing an Indication With a Market Quote

A hypothetical reinsurer models $R$, in thousands of dollars, with distribution $P(R=0)=0.80$, $P(R=500)=0.15$, and $P(R=2{,}000)=0.05$. Its pricing assumptions are $k=0.20$, fixed expenses of $10$ thousand dollars, brokerage of $5\%$, and target profit of $4\%$ of gross premium. A market quote is $325$ thousand dollars. Calculate the technical indication and, if the quote is accepted with these assumptions, the residual after expected recovery, risk load, fixed expenses, and brokerage. Round to two decimals in thousands. The method is to calculate the indication first, then reconcile the actual quote on the same basis. It avoids calling the quote’s entire difference from expected recovery “profit.”

Expected recovery. $E[R]=(0.80)(0)+(0.15)(500)+(0.05)(2{,}000)=0+75+100=175$ thousand dollars.

Risk load. $E[R^2]=(0.15)(500^2)+(0.05)(2{,}000^2)=37{,}500+200{,}000=237{,}500$. Thus $\operatorname{Var}(R)=237{,}500-(175)^2=237{,}500-30{,}625=206{,}875$. The standard deviation is $\sqrt{206{,}875}=454.84$ thousand dollars, so $L=(0.20)(454.84)=90.97$ thousand dollars.

Technical indication. The dollar provisions total $175+90.97+10=275.97$. Brokerage and target profit total $5\%+4\%=9\%$, giving $G=275.97/(1-0.09)=303.26$ thousand dollars.

Market-quote reconciliation. At the $325$-thousand-dollar quote, brokerage is $0.05(325)=16.25$. The residual after the specified recovery, risk load, fixed expense, and brokerage is $325-175-90.97-10-16.25=32.78$ thousand dollars. If all assumptions hold, this residual is available for profit and any other unmodeled items. The target profit provision at the quote is $0.04(325)=13.00$, so the residual exceeds that target by $32.78-13.00=19.78$ thousand dollars. A market quote is not automatically a technical indication: the difference is interpretable only relative to the stated assumptions and components.

Common Traps and Interpretation

Mixing gross and net bases. If brokerage is a percentage of gross premium, include it in the gross-up denominator. Do not calculate it as a percentage of expected recovery unless that is explicitly the contract or pricing convention.
Using a risk load without naming its basis. A load such as $k$ times standard deviation requires a stated $k$ and a defined recovery distribution. It is not the same as an arbitrary percentage of expected recovery.
Counting an item twice. If a provision already includes an expense or uncertainty allowance, adding the same item separately inflates the price.
Calling every difference profit. A market quote can reflect assumptions or provisions not in the simplified model. Reconcile only the components that have been specified.

A component view is useful even when the final quote is negotiated rather than mechanically calculated. It clarifies how sensitive an indication is to expected recoveries, the risk allowance, and the pricing bases for expenses and brokerage. It also makes comparisons more meaningful: two premiums are not directly comparable if one includes a percentage expense or risk provision that the other omits.

Key takeaway: State each provision and its calculation basis before solving for premium. Add dollar components directly, gross up percentage-of-premium components, and treat a market quote as distinct from the technical indication unless both are built on the same assumptions.

Check Your Understanding

For each problem, calculate on the stated basis and keep dollar amounts in thousands unless noted otherwise.

A hypothetical treaty recovery has distribution $P(R=0)=0.60$, $P(R=200)=0.30$, and $P(R=1{,}000)=0.10$. Use a standard-deviation load with $k=0.10$, fixed expenses of $8$, brokerage of $5\%$, and target profit of $5\%$ of gross premium. Find the gross technical indication to two decimals. (Answer: $219.33$ thousand dollars.)
A cover has expected recovery of $450$ and a separately specified risk load of $75$. Fixed expenses are $20$; variable expenses are $2\%$, brokerage is $8\%$, and target profit is $4\%$, each as a percentage of gross premium where applicable. Calculate the gross indication to two decimals. (Answer: $633.72$ thousand dollars.)
A quote is $450$. The pricing basis has expected recovery $300$, risk load $60$, fixed expenses $15$, brokerage $6\%$ of gross premium, and a target profit of $5\%$ of gross premium. Calculate the technical indication and the residual after recovery, risk load, expenses, and brokerage at the quote. (Answer: technical indication $421.35$ thousand dollars; residual $48.00$ thousand dollars.)
A hypothetical program has expected recovery $600$, risk load $100$, and fixed expenses $50$, all in thousands. Brokerage is $12\%$ and target profit is $9\%$ of gross premium. Calculate the gross technical indication to two decimals and identify the gross-up denominator. (Answer: $949.37$ thousand dollars; denominator $0.79$.)

What You'll Learn