Tutorials › Exam ALTAM - Advanced Long-Term Actuarial Mathematics › Flexible Premiums and Reserve Adequacy

Exam ALTAM · Universal Life Profit and Reserves · Tutorial 638 of 999

Flexible Premiums and Reserve Adequacy

Learn to project flexible premiums through policy charges and crediting, then assess the resulting insurer-level reserve on a clearly specified valuation basis.

Advanced 12 min read

What You'll Learn

  • Project account values when premiums vary by amount and timing
  • Calculate premium loads, cost-of-insurance charges, and net amount at risk
  • Compare different premium schedules with the same total contribution
  • Separate the policy account value from the insurer’s supplementary reserve
  • Calculate prospective reserves using survival, mortality, interest, expenses, and future charges
  • Identify timing errors that distort account projections and reserve results

Why Premium Flexibility Changes the Valuation

A universal life policyholder may vary the amount and timing of premiums, subject to the contract’s requirements. Consequently, a premium schedule is not just a payment detail: it changes how much is credited to the policy account, the amount exposed to future charges, and the future insurer cash flows used in a reserve calculation. Two schedules with the same total premiums can produce different account values and different mortality margins.

The calculations below use a deliberately specified annual model. It is a valuation framework for examining premium sensitivity, not a claim that every universal life contract applies charges in this exact sequence. All policies and numerical assumptions are hypothetical. Premiums are paid at the start of each year; the policy is assumed to remain in force; and the credited interest rate and mortality rates are deterministic. There are no surrender benefits, taxes, guarantees, or changes of valuation basis in the examples.

Definition: Let \(A_k\) be the account value at the start of policy year \(k\), before that year’s premium is paid. Let \(P_k\) be the premium paid at that time, and let \(\lambda\) be the proportion deducted as a premium load. After the premium, an expense charge \(F_k\) and cost-of-insurance charge \(C_k\) are deducted. If the annual credited rate is \(j\), the account value at the start of the next year is $$ A_{k+1}=\bigl[A_k+(1-\lambda)P_k-F_k-C_k\bigr](1+j). $$

In this model, the death benefit is \(\max(B,A_{k+1})\), where \(B\) is a specified minimum benefit. The account value is available to fund the account-based part of that benefit; the insurer’s additional death-benefit exposure is therefore \(\max(B-A_{k+1},0)\). For the cost-of-insurance charge, define the net amount at risk before that charge as \(N_k=\max\{B-[A_k+(1-\lambda)P_k-F_k],0\}\). If the annual mortality rate is \(q_k\), the charge is \(C_k=q_kN_k\). This is a stated charge convention for the tutorial; a contract may define its charge base or timing differently.

Project the Premium Schedule Before Valuing It

For a flexible-premium policy, the account projection must use the premium actually assumed at each payment date. Do not replace a varying schedule with an average premium unless the valuation model explicitly justifies that substitution. The sequence matters: the premium load is deducted from each payment, while the account value left after charges earns the credited rate for the year.

Worked Example: Projecting Two Unequal Premiums

A hypothetical policy has \(A_0=0\), minimum death benefit \(B=\$100\), a 10% premium load, a \(\$1\) annual expense charge, annual mortality rate \(q=0.02\), and annual credited rate \(j=0.05\). The policyholder pays \(\$20\) at time 0 and \(\$10\) at time 1. Find the account value at time 2 and the cost-of-insurance charge in each year. Use the charge convention just defined.

Method. Project year by year, calculating the net amount at risk from the account immediately before the cost-of-insurance charge. This avoids applying the mortality rate to the wrong account value.

Year 1. The premium after its load is \(\$20(1-0.10)=\$18\). After the expense charge, the balance before the cost-of-insurance charge is \(\$0+\$18-\$1=\$17\). Thus \(N_0=\$100-\$17=\$83\), and \(C_0=0.02(\$83)=\$1.66\). The amount credited with interest is \(\$17-\$1.66=\$15.34\). Therefore \(A_1=\$15.34(1.05)=\$16.107\).

Year 2. The premium after its load is \(\$10(0.90)=\$9\). The balance after the premium and expense charge is \(\$16.107+\$9-\$1=\$24.107\). Hence \(N_1=\$100-\$24.107=\$75.893\), and \(C_1=0.02(\$75.893)=\$1.51786\). The amount credited with interest is \(\$24.107-\$1.51786=\$22.58914\). Thus \(A_2=\$22.58914(1.05)=\$23.718597\).

The year 1 and year 2 cost-of-insurance charges are \(\$1.66\) and \(\$1.51786\); the year 2 opening account value is \(\$23.718597\). Rounded to cents only at the end, the charges are \(\$1.66\) and \(\$1.52\), and the time 2 account value is \(\$23.72\). As a check, the time 2 value is the year 2 post-charge balance, \(\$22.58914\), multiplied by \(1.05\).

Why Payment Timing Matters

A premium paid earlier can earn credited interest for longer and can reduce the net amount at risk earlier. Those effects usually support a larger later account value, but the precise outcome depends on the charges and contract rules. In particular, a larger account may reduce future cost-of-insurance charges under a net-amount-at-risk charging method. That reduction is not the same as a premium load: the load is deducted from the premium, whereas the cost-of-insurance charge depends on the specified risk exposure.

Worked Example: Same Total Premium, Different Timing

Use the same hypothetical assumptions as the preceding example, except that the policyholder has a total of \(\$30\) to pay over two years. Compare schedule A, \(\$30\) at time 0 and \(\$0\) at time 1, with schedule B, \(\$15\) at each time. Find the time 2 account value under each schedule. The initial account value is zero.

Method. Apply the same charges, mortality rate, and crediting rate to both schedules. Do not compare only the total premium: each schedule has its own year-by-year account and net amount at risk.

Schedule A, year 1. After the 10% load and \(\$1\) expense charge, the balance is \(\$27-\$1=\$26\). The cost-of-insurance charge is \(0.02(\$100-\$26)=\$1.48\). Therefore \(A_1=(\$26-\$1.48)(1.05)=\$25.746\).

Schedule A, year 2. With no premium, the pre-mortality-charge balance is \(\$25.746-\$1=\$24.746\). The charge is \(0.02(\$100-\$24.746)=\$1.50508\). Therefore \(A_2=(\$24.746-\$1.50508)(1.05)=\$24.402966\).

Schedule B, year 1. The premium after its load is \(\$13.50\), so the balance after the expense charge is \(\$12.50\). The cost-of-insurance charge is \(0.02(\$100-\$12.50)=\$1.75\). Thus \(A_1=(\$12.50-\$1.75)(1.05)=\$11.2875\).

Schedule B, year 2. The balance after the second net premium and expense charge is \(\$11.2875+\$13.50-\$1=\$23.7875\). The cost-of-insurance charge is \(0.02(\$100-\$23.7875)=\$1.52425\). Thus \(A_2=(\$23.7875-\$1.52425)(1.05)=\$23.3764125\).

The time 2 values, rounded to cents, are \(\$24.40\) for schedule A and \(\$23.38\) for schedule B. Schedule A ends with an account value \(\$1.03\) higher, even though both schedules pay \(\$30\) in total. The result reflects earlier crediting and the different sequence of charges; it is not a general guarantee that front-loading always produces a particular advantage. A common mistake is to apply two years of interest to the total premium in either schedule. That ignores the premium load, expense deductions, and mortality charges.

Prospective Reserve on a Specified Basis

An account value and an insurer reserve answer different questions. The account value is the policyholder’s account under the contract’s projection rules. A supplementary insurer-level reserve can be defined for the expected net cost of the additional death-benefit exposure and expenses, after allowing for future insurer income from specified charges. Under this approach, the account value is projected separately and is not counted again as premium income or as a second reserve asset.

For precision, suppose valuation is at the start of year \(t\), future policy-year cash flows are projected while the policy remains in force, and actual expenses \(E_k\), expense charges \(F_k\), premium loads \(\lambda P_k\), and cost-of-insurance charges \(C_k\) occur at the start of year \(k\). A death claim for the additional benefit occurs at year end. Let \(q_k\) be the probability of death during year \(k\), conditional on being in force at its start, and let \({}_{k-t}p_t\) be the probability of surviving from the valuation time to the start of year \(k\). At annual valuation interest rate \(i\), set \(v=1/(1+i)\). The prospective reserve for these net insurer cash flows is

$$ R_t=\sum_{k=t}^{n-1} {}_{k-t}p_t \left[ v^{k-t}(E_k-\lambda P_k-C_k-F_k) +v^{k-t+1}q_k\max(B-A_{k+1},0) \right]. $$

This formula assumes the projected premium schedule and charges are known, the annual cash-flow timing just stated applies, and the policy persists for the projection period unless death occurs. It is a net reserve on this defined basis; a negative result indicates projected income exceeds projected net outgo. Whether a valuation framework imposes a floor is a separate requirement and is not assumed here.

1
Project the account.
Use each scheduled premium and the stated order of premium load, expense charge, cost-of-insurance charge, and crediting.
2
Identify insurer cash flows.
At the start of a year, subtract premium loads and charges from actual expenses. At year end, include the expected additional death-benefit claim.
3
Weight and discount.
Multiply each future year’s cash flow by the probability the policy reaches that year, then discount it using the stated valuation interest rate and payment timing.

Why the prospective formula holds. The reserve is the expected present value of future insurer outgo less the expected present value of future insurer income. For a cash flow at the start of year \(k\), the policy must first survive to that date, giving the factor \({}_{k-t}p_t\), and the cash flow is discounted \(k-t\) years. For the additional death claim at the end of that year, death must occur conditional on reaching the start, giving \(q_k\), and the claim is discounted one more year. Adding these mutually timed expected net cash flows gives the displayed sum. This is an application of linearity of expectation; it does not require independence between future cash flows.

Worked Example: Testing a One-Year Reserve

A hypothetical policy is valued at time 0 with \(A_0=\$20\), \(B=\$100\), and a single future premium \(P_0=\$10\). The premium load is 10%, the expense charge is \(\$1\), and the cost-of-insurance charge is \(qN_0\), with \(q=0.02\). The credited rate is 4%. Actual insurer expense at time 0 is \(\$1.20\), and the expense charge received by the insurer is \(\$1\). The valuation interest rate is \(i=4\%\), and the death claim is paid at year end. Calculate the one-year supplementary net reserve.

Method. First project the account through the year, then calculate the expected additional claim and discount it. Finally, combine that claim cost with the start-of-year expenses and charge income. The account value is kept separate from this supplementary reserve.

The net premium after load is \(\$10(0.90)=\$9\). After the \(\$1\) expense charge, the pre-mortality-charge balance is \(\$20+\$9-\$1=\$28\). Thus \(N_0=\$100-\$28=\$72\) and \(C_0=0.02(\$72)=\$1.44\). The amount credited with interest is \(\$28-\$1.44=\$26.56\), so \(A_1=\$26.56(1.04)=\$27.6224\).

The additional death claim at year end is \(\$100-\$27.6224=\$72.3776\). Its expected present value is \(0.02(\$72.3776)/1.04=\$1.391876923\). At time 0, actual expense is \(\$1.20\); income consists of the \(\$1\) premium load, \(\$1.44\) cost-of-insurance charge, and \(\$1\) expense charge. Therefore \(R_0=\$1.391876923+\$1.20-\$1-\$1.44-\$1=-\$0.848123077\).

Rounded to cents, the supplementary net reserve is \(-\$0.85\). The negative sign means that, on this one-year basis, the specified charge income exceeds the expected additional claim and expense outgo in present value. It is not a conclusion that the account value is negative: the projected account at year 1 is \(\$27.62\). The check is to keep the two measures separate and include the death claim at year end, not at time 0.

Common Errors and Practical Checks

Flexible-premium projections are particularly sensitive to timing. A calculation can use the right premium total and the right annual interest rate yet still be wrong if it credits the premium too early or computes charges on the wrong balance. Before accepting a result, check both the account recursion and the cash-flow valuation basis.

Common mistake—charging mortality before applying the premium. In Worked Example 1, the incorrect calculation \(C_0=0.02(\$100-\$0)=\$2.00\) uses the opening account before the time 0 premium. Under the stated convention, first apply the premium load and expense charge: the pre-charge balance is \(\$17\). The correct charge is therefore \(0.02(\$100-\$17)=\$1.66\). The wrong order overstates the charge by \(\$0.34\) and understates the projected account value.

Other useful checks include verifying that the premium load applies to the premium, not to the entire account; confirming that credited interest applies only after the specified deductions; and distinguishing the mortality rate used to calculate a charge from the timing of the expected death claim in a reserve. A prospective reserve also needs a clearly specified future premium schedule. If premiums are uncertain, a deterministic schedule is an assumption for the valuation, not a consequence of the policy’s flexibility.

The central discipline is to state the contract mechanics and valuation assumptions before calculating. Then project the account consistently, identify which cash flows belong to the insurer, and discount each at its proper time. These methods are aligned with the SOA Exam ALTAM (Advanced Long-Term Actuarial Mathematics) syllabus.

Check Your Understanding

Use the stated annual timing and charge conventions. Round monetary answers to cents only at the end.

  1. A hypothetical policy has \(A_0=\$5\), \(P_0=\$12\), premium load 10%, expense charge \(\$0.50\), \(B=\$50\), mortality rate \(q=0.01\), and credited rate \(j=0.04\). Find the cost-of-insurance charge and \(A_1\). (Answer: charge \(\$0.347\); \(A_1=\$15.55\).)
  2. For a one-year supplementary reserve, take \(A_0=\$10\), \(P_0=\$20\), premium load 5%, expense charge \(\$0.50\), actual expense \(\$0.80\), \(B=\$60\), \(q=0.01\), and both credited and valuation interest rates of 3%. Find the net reserve, including the premium load, cost-of-insurance charge, and expense charge as insurer income at time 0. (Answer: \(-\$0.71\).)
  3. Under the assumptions in Worked Example 2, what are the time 2 account values for a \(\$30\) premium at time 0 and no premium at time 1, versus \(\$15\) at each time? State which schedule produces the larger value. (Answer: \(\$24.40\) versus \(\$23.38\); the time 0 payment schedule is larger.)
  4. A one-year valuation has an additional death claim of \(\$40\) if death occurs at year end, mortality rate \(q=0.03\), valuation interest rate \(i=0.05\), actual expense of \(\$2\) at time 0, and total charge income of \(\$0.50\) at time 0. With no other cash flows, calculate the supplementary net reserve. (Answer: \(\$2.64\).)
  5. Explain why two premium schedules with the same total premium need not produce the same prospective reserve, even when mortality and interest assumptions are unchanged. (Answer: Premium timing changes the account projection and therefore future net amounts at risk and charges; it also changes the timing of premium-load income and its discounting.)
Clinical Biostats