Acquisition Expenses And Reserve Recognition

Why Acquisition Expenses Matter to Reserves

An insurer may incur substantial costs when a contract is issued, while premiums and benefits emerge over several years. A reserve calculation must therefore make its expense assumptions explicit: an expense paid at issue is not the same cash flow as an expense due at a later anniversary. The timing matters because a prospective reserve values future cash flows from the valuation date; it does not automatically carry past expenses forward.

This tutorial considers hypothetical annual-premium contracts with annual benefits and explicit expense cash flows. The focus is how acquisition expenses are recognized in the valuation and how that recognition affects reserves at issue and at early durations. The examples use a simple one-state model: a policy is either in force or terminated by death. The annual mortality probability and interest rate are constant, and all cash flows occur at the stated times.

Definition: Let $V_k$ be the prospective reserve at time $k$, immediately before the premium and any expense due at that time. For an annual-premium contract, suppose a premium $P$ is received at times $k=0,1,\ldots,n-1$ while the policy is in force; a death benefit $B$ is paid at the end of a policy year if death occurs during that year; and an expense $e_k$ is incurred at time $k$. With annual effective interest rate $i$, $v=1/(1+i)$, one-year death probability $q$, and survival probability $p=1-q$, the reserve satisfies $$ V_k=-P+e_k+v\left(qB+pV_{k+1}\right),\qquad k=0,\ldots,n-1. $$ If there are no cash flows after the contract term, $V_n=0$. The reserve includes cash flows due at time $k$; a cash flow already paid before time $k$ is not included.

The equation uses a consistent timing convention. At time $k$, the premium is an inflow and the expense is an outflow, so their net contribution to the liability is $-P+e_k$. During the following year, the death benefit is paid at year-end with probability $q$. If the policyholder survives, the contract continues and the reserve at time $k+1$ is relevant. Both end-of-year amounts are discounted by $v$.

How the Issue Expense Enters the Reserve

An acquisition expense paid at issue is represented by $e_0$. A renewal expense due at a later anniversary is represented by the corresponding $e_k$. These are separate assumptions, even if the insurer describes both as expenses per policy. The issue expense is included in $V_0$, because it is due at time 0 under the definition above. Once time 0 has passed, it is a past cash flow and is not added again to the prospective reserve.

That does not mean that the issue expense is irrelevant to the contract’s overall financial results. It may be funded through the premium charged, and it affects the insurer’s cash flow at issue. But a prospective reserve at a later duration is based on the remaining future cash flows and the valuation basis at that duration. A method that allocates acquisition costs across future durations would require a further reserve-basis rule; it should not be silently substituted for the explicit-expense model here.

Worked Example: Changing Only the Issue-Expense Assumption

A hypothetical two-year term contract has $B=\$10{,}000$, annual death probability $q=0.10$, annual effective interest rate $i=0.05$, and a level annual premium $P=\$952.38$, payable at issue and at time 1 if the policy is in force. A renewal expense of $\$50$ is due at time 1. Compare two valuation assumptions for the acquisition expense due at issue: $\$300$ and $\$450$. Find $V_1$ and $V_0$ under each assumption.

Method. Apply the prospective reserve recursion from the end of the contract backward. This isolates the effect of changing the issue expense while keeping the premium and all later cash flows fixed.

The survival probability is $p=1-0.10=0.90$, and the discount factor is $v=1/1.05=0.95238095$. At time 2, the contract has ended, so $V_2=0$.

At time 1, the premium and renewal expense are due immediately. If death occurs in the next year, the benefit is paid at time 2; otherwise the terminal reserve is zero. Thus

V_1=-952.38+50+\frac{1}{1.05}(0.10\times 10{,}000+0.90\times 0).

The discounted death-benefit amount is $0.95238095\times 1{,}000=\$952.38$. Therefore

V_1=-952.38+50+952.38=\$50.00.

With the $\$300$ issue expense, the time-0 reserve is

V_0=-952.38+300+0.95238095(0.10\times 10{,}000+0.90\times 50) =\$342.86.

In detail, the survival-contingent time-1 reserve contributes $0.90\times 50=\$45.00$. The amount inside the discounted term is therefore $\$1{,}000+\$45=\$1{,}045$, whose present value is $\$995.24$. Hence $V_0=-\$952.38+\$300+\$995.24=\$342.86$.

With a $\$450$ issue expense, only the time-0 expense term changes:

V_0=-952.38+450+995.24=\$492.86.

The check is direct: increasing the issue expense by $\$150$ increases $V_0$ by $\$150$, while $V_1$ remains $\$50.00$ because the issue expense has already occurred by time 1. A common mistake is to add the issue expense to $V_1$ as well; that counts a past cash flow a second time.

Deriving the Reserve Recursion

The recursion follows from conditioning on what happens during the next policy year. At time $k$, first account for the premium and expense due immediately. Then consider the end-of-year outcomes. Death, with probability $q$, produces the benefit $B$; survival, with probability $p$, leads to the reserve $V_{k+1}$. The expected end-of-year liability is therefore $qB+pV_{k+1}$, and its value at time $k$ is discounted by $v$.

V_k=(-P+e_k)+v\left(qB+pV_{k+1}\right).

This is a prospective equivalence statement: the reserve is the present value of future outgo less the present value of future income, with current-time cash flows included according to the stated convention. The derivation assumes level annual mortality and interest for the example formula, premiums payable only while the policy is in force, and no other cash flows unless explicitly added. If premiums are paid at a different time, or if a benefit is paid at a different time, the cash-flow timing and discounting must be changed accordingly.

The recursion also makes the local effect of an expense assumption transparent. If two valuation calculations use identical benefits, premiums, mortality, interest, and later expenses, but use issue expenses $e_0$ and $e_0+\Delta$, then their time-0 reserves differ by $\Delta$. At a later time $k>0$, the past issue expense is absent from the prospective cash flows, so it does not directly change $V_k$. This comparison presumes that changing the expense assumption does not also change the premium or another valuation assumption.

Gross Premiums and Expense Recognition

An expense assumption can affect reserves through more than a direct expense term. If the premium is determined to fund expected benefits and expenses, changing the assumed acquisition expense may change the premium itself. That is a different comparison from holding the contractual premium fixed and changing only the reserve expense assumption. Always identify which quantity is fixed in the problem.

For a level gross premium determined by the equivalence principle, the expected present value of premiums equals the expected present value of benefits plus expenses, using the same mortality, interest, timing, and policy-continuation assumptions. For a term contract with annual premiums due at the beginning of each year while the policy is in force, this equality can be written

P\sum_{k=0}^{n-1}v^k\,{}_kp_x = \operatorname{EPV}(\text{benefits})+\operatorname{EPV}(\text{expenses}),

where ${}_kp_x$ is the probability that a life aged $x$ survives $k$ years. The issue expense at time 0 is not discounted; an expense due at a later time is discounted and is weighted by the probability that the policy is then in force, if it is conditional on being in force.

Worked Example: Finding a Gross Premium With an Issue Expense

Use the two-year contract and assumptions from the preceding example: $B=\$10{,}000$, $q=0.10$, $p=0.90$, $i=0.05$, and $v=20/21$. Suppose the issue expense is $\$300$, the expense at time 1 is $\$50$ if the policy is in force, and the level premium $G$ is to be set by the equivalence principle. Find $G$, then calculate the prospective reserves at times 1 and 0.

Method. First equate expected present values at issue. Then substitute the resulting premium into the reserve recursion. This avoids treating the issue expense as if it were a renewal expense.

The expected present value of the death benefit is

10{,}000(0.10)v+10{,}000(0.90)(0.10)v^2 =952.38095+816.32653 =\$1{,}768.71.

The expected present value of expenses is the issue expense plus the time-1 expense weighted for survival and discounted:

300+0.90(50)v=300+42.85714=\$342.86.

The expected present value of one dollar of level premiums is $1+0.90v=1+0.85714286=13/7$. The first premium is certain at issue; the second is payable only if the policyholder survives to time 1. Therefore

G=\frac{1{,}768.70748+342.85714}{13/7} =\frac{310{,}400}{273} =\$1{,}136.99634.

The time-1 reserve includes the time-1 premium and renewal expense, as well as the death benefit payable in the following year:

V_1=-1{,}136.99634+50+0.95238095(0.10\times10{,}000) =-\$134.62.

At issue, the recursion gives

V_0=-1{,}136.99634+300 +0.95238095\left(1{,}000+0.90(-134.61538)\right) =\$0.00.

The zero at issue checks the gross-premium equivalence calculation: the premium was chosen so that expected present value of premiums equals expected present value of benefits and expenses. The negative time-1 prospective reserve is also consistent with the assumptions: the remaining premium exceeds the present value of the remaining benefit and renewal expense. This calculation does not impose a minimum-reserve floor; any such requirement would be an additional rule, not part of the stated prospective equivalence calculation.

Expense Timing: A Frequent Source of Error

The amount of an expense alone does not determine its reserve effect. Its payment time and whether it is conditional on the policy remaining in force both matter. An issue cost is certain at time 0 under the model above. A renewal expense due at time 1 only if the policy is still in force has expected present value $vp e_1$ at time 0, not $ve_1$ and not $pe_1$. The discount factor accounts for interest, while $p$ accounts for the chance the expense is incurred.

Expense assumption	Time-0 expected present value	Reason
Issue expense $e_0$, paid at time 0	$e_0$	It is certain and due immediately
Expense $e_1$, due at time 1 only if in force	$vpe_1$	It is discounted and conditional on survival
Expense $e_1$, due at time 1 regardless of survival	$ve_1$	It is discounted but not conditional on survival

For example, with $i=0.05$, $p=0.90$, and a $\$50$ expense due at time 1 only if in force, the expected present value is $0.95238095\times0.90\times\$50=\$42.86$. If the expense is due regardless of survival, the expected present value is $0.95238095\times\$50=\$47.62$. The difference is not a change in discounting; it is the probability that the conditional expense is incurred.

Worked Example: Correcting an Expense-Timing Calculation

A hypothetical policy has a $\$200$ expense due at time 1 only if the policyholder survives the year. The annual effective interest rate is $0.04$, and the one-year survival probability is $0.95$. A candidate values the expense at time 0 as $v(200)=\$192.31$. Find the correct expected present value and the amount of the overstatement.

Method. Multiply by the probability that the policy is in force when the expense is due, then discount one year.

Here $v=1/1.04=0.96153846$. The expense is incurred only on survival, so its expected present value is

0.95(200)(0.96153846)=\$182.69.

The incorrect calculation assigns probability 1 to an expense that is conditional on survival. Its result is $\$200(0.96153846)=\$192.31$, so the overstatement is $\$192.31-\$182.69=\$9.62$, to the nearest cent. The check is that the correct amount is $0.95$ times the incorrect amount: $0.95\times\$192.30769=\$182.69231$. Treating an in-force expense as unconditional is the error.

Practical Checks for Reserve Recognition

Before calculating, write down the reserve timing convention and make a cash-flow timeline. Mark whether each premium, expense, and benefit is due at the valuation time, during the year, or at year-end. Then identify whether the cash flow is conditional on survival or another state. These steps prevent an expense from being placed in the wrong duration or multiplied by the wrong probability.

Set the valuation basis.
State the interest rate, mortality or survival probabilities, premium amount, benefit amount, and expense assumptions. Distinguish contractual cash flows from valuation assumptions.

Locate every cash flow.
Record its payment time and whether it is certain, conditional on survival, or conditional on another event.

Use the prospective reserve definition.
Include cash flows due at the valuation time under the chosen convention, and do not reinsert expenses already paid.

Check the result.
Verify the recursion against expected present values, and check that changing only the issue expense changes the issue reserve by the same amount.

A reserve may be negative under an unconstrained prospective calculation, as in the gross-premium example. That result can arise when future premiums exceed the present value of future benefits and expenses. Do not change the calculation merely because the result is negative; first check the cash-flow timing and assumptions. If a problem specifies a reserve floor or a separate statutory treatment, apply that condition explicitly.

Key takeaway: Under an explicit prospective expense basis, an acquisition expense due at issue enters the reserve at issue as a current outgo. At later durations it is a past cash flow and is not counted again. Changing the premium to fund that expense is a separate effect that must be included only when the premium assumption is also being changed.

Check Your Understanding

For each problem, use the stated timing convention and round money to the nearest cent.

A one-year term policy has $B=\$5{,}000$, $q=0.04$, $i=0.03$, and a premium of $\$200$ due at time 0. An issue expense of $\$120$ is due at time 0, and there are no other expenses. Find $V_0$. (Answer: $\$114.17$)
Two valuations use the same contract cash flows and premium, but the second assumes an issue expense $\$75$ higher than the first. What is the difference in their time-0 prospective reserves, and what is the direct difference in their time-1 prospective reserves if all later assumptions agree? (Answer: Time-0 reserve is $\$75.00$ higher; time-1 difference is $\$0.00$)
A one-year policy has $B=\$1{,}000$, $q=0.10$, $i=0.05$, premium $P=\$100$, and an issue expense of $\$40$. There are no later expenses. Calculate $V_0$. (Answer: $\$35.24$)
A one-year policy has $B=\$2{,}000$, $q=0.05$, $i=0.04$, and a certain issue expense of $\$30$. Find the level premium set by the equivalence principle, with no other expenses. (Answer: $\$126.15$)
A $\$120$ expense is due at time 1 only if the policyholder survives. The annual effective interest rate is $0.05$ and the one-year survival probability is $0.90$. Find its expected present value at time 0. (Answer: $\$102.86$)

Expense assumption	Time-0 expected present value	Reason
Issue expense \(e_0\), paid at time 0	\(e_0\)	It is certain and due immediately
Expense \(e_1\), due at time 1 only if in force	\(vpe_1\)	It is discounted and conditional on survival
Expense \(e_1\), due at time 1 regardless of survival	\(ve_1\)	It is discounted but not conditional on survival

What You'll Learn