Introduction
A participating policy can produce a surplus in a year without that surplus appearing immediately as shareholder profit. The insurer may instead allocate part of it to discretionary policyholder benefits. The timing and amount of that allocation affect when profit emerges, even when the contractual guarantees are unchanged.
The key distinction is between a guaranteed benefit, which is part of the policy’s contractual floor, and a discretionary benefit, whose declaration or amount is determined by the insurer under the policy terms. This tutorial considers a deterministic accounting projection: it tracks annual profit after guaranteed benefits and expenses, then allows for discretionary benefit payments and the reserve held for benefits already declared.
The reserve \(L_t\) is a liability measured at the end of the year on the stated valuation basis. If a bonus is declared at the end of year \(t\) and is payable one year later, its reserve is the present value at \(t\) of that payment. A bonus paid immediately is included in \(C_t\), not held in \(L_t\). The \(S_t\) definition includes investment income on assets backing the liabilities, so the formula uses the actual change in reserve, \(L_t-L_{t-1}\), rather than adding a separate interest adjustment to that change.
In the examples, the projection is hypothetical and deterministic. The policy remains in force throughout the stated projection, with no death, lapse, or other decrement. The stated annual effective interest rate is used to value declared bonuses payable later. This isolates the effect of allocation timing; it is not a mortality model.
Why Allocation Timing Changes Profit Emergence
A declared bonus payable in the future creates a liability now. Consequently, the year of declaration bears the increase in the bonus reserve, even though no cash has yet been paid. When the bonus is later paid, the liability is released, offsetting part of the cash outgo in that later year’s profit calculation.
The total-profit reconciliation follows directly by summing annual profit. The reserve movements telescope: every intermediate reserve appears once as an increase and once as a decrease.
If the projection starts and ends with no outstanding declared-benefit liability, total profit equals total pre-allocation net results less total discretionary payments. Moving a fixed payment to a different declaration date can change the annual pattern of profits without changing that reconciliation. Increasing the total bonus payment, by contrast, reduces total profit by the additional amount, all else equal. This accounting identity does not say that two different benefit patterns have the same value to policyholders; it isolates the profit effect under the stated basis.
Use \(S_t\) as given or calculate it from the projection, keeping guaranteed benefits and reserve movements separate from discretionary benefits.
Enter payments in \(C_t\) in the year they are paid, not the year they were first considered.
Calculate the closing reserve \(L_t\) on the stated basis, then find its change from \(L_{t-1}\).
Apply \(\pi_t=S_t-C_t-(L_t-L_{t-1})\), then check the total against the telescoping reconciliation.
Worked Applications
The examples use a simplified annual profit projection. All amounts are in dollars, and the stated \(S_t\) values are already net of guaranteed benefits and other items described in the definition. A change in a discretionary allocation does not alter the contractual guaranteed benefits in these examples.
Worked Example: Declaring a Bonus Earlier
A hypothetical policy has a guaranteed benefit that is unchanged under either allocation. The annual effective interest rate is \(5\%\). There is no mortality or lapse during the two-year projection; all values are deterministic. Before discretionary bonuses, the net results are \(S_1=50\) and \(S_2=50\). Compare (A) declaring a discretionary bonus of \(30\) at the end of year 1, payable at the end of year 2, with (B) declaring and paying a terminal bonus of \(30\) at the end of year 2. Calculate annual and total profit under each approach.
Method. For each year, identify the payment and the reserve for declared but unpaid benefits. In case A, the year-1 reserve is the present value of the year-2 payment:
In year 1, no discretionary cash is paid, so \(C_1=0\), and \(L_0=0\). Therefore:
At the end of year 2 the bonus is paid, so \(C_2=30\), and the unpaid-benefit reserve falls from \(28.5714\) to zero. Thus:
Total profit in case A is \(21.4286+48.5714=70.0000\). In case B, there is no declared unpaid bonus at the end of year 1, so \(L_1=0\) and \(\pi_1=50\). At the end of year 2, the bonus is declared and paid immediately: \(C_2=30\), \(L_2=L_1=0\), and \(\pi_2=50-30=20\). Total profit is also \(50+20=70\).
The early declaration moves profit emergence from \((50,20)\) to approximately \((21.43,48.57)\), while the total remains \(70\). The common mistake is to charge the full \(30\) as a year-1 liability, giving year-1 profit of \(20\). That uses the payment amount instead of its reserve value. The correction is to discount the year-2 payment at \(5\%\), giving \(L_1=28.5714\).
Check. Both cases have total pre-allocation results of \(100\), total bonus payments of \(30\), and zero ending liability. The reconciliation gives \(100-30=70\), matching both calculations.
Worked Example: Changing the Amount Allocated
A hypothetical two-year policy remains in force throughout the projection, with no mortality or lapse. The annual effective interest rate is \(4\%\), and the deterministic pre-allocation results are \(S_1=60\) and \(S_2=60\). In either design, a bonus is declared at the end of year 1 and paid at the end of year 2. Compare a bonus of \(20\) with a bonus of \(35\). The guaranteed benefit is identical in both designs.
Method. For a bonus \(B\), the end-of-year-1 reserve is \(B/1.04\). With \(B=20\):
In year 2, the payment is \(20\) and the closing reserve is zero:
Total profit is \(40.7692+59.2308=100.0000\), also obtained as \(120-20=100\).
With \(B=35\), the reserve at the end of year 1 is:
The year-2 profit is:
Total profit is \(26.3462+58.6538=85.0000\), matching \(120-35=85\). Increasing the discretionary bonus by \(15\) lowers total profit by \(15\). It also changes the year-by-year pattern: the larger reserve reduces year-1 profit, while the larger payment and reserve release both affect year-2 profit. The guaranteed benefit remains unchanged; the difference is in the discretionary allocation.
Check. Each design starts and ends with no unpaid-bonus reserve. Total profit must therefore equal total pre-allocation results less the bonus paid. The two reconciliations are \(120-20=100\) and \(120-35=85\).
Worked Example: Allocating a Share of Each Year’s Surplus
A hypothetical policy remains in force for three years, with no mortality or lapse. The annual effective interest rate is \(4\%\). The deterministic pre-allocation results are \(S_1=30\), \(S_2=70\), and \(S_3=0\). At each of the ends of years 1 and 2, the insurer declares a bonus equal to \(50\%\) of that year’s result. Each declared bonus is payable one year after declaration. Calculate annual profit and reconcile the total. The contractual guaranteed benefits are not changed by these discretionary declarations.
Method. The bonuses declared at the ends of years 1 and 2 are \(0.50(30)=15\) and \(0.50(70)=35\), respectively. At the end of year 1, the \(15\) payable in year 2 has reserve:
No bonus is paid in year 1, so:
In year 2, the first bonus of \(15\) is paid. The second bonus of \(35\) is declared and payable in year 3, so the closing reserve is:
The liability change is \(33.6538-14.4231=19.2307\), using rounded values. Therefore:
In year 3, the \(35\) is paid and the reserve falls to zero:
Using unrounded values, total profit is \(15.576923+35.769231-1.346154=50.000000\). This equals total pre-allocation results \(30+70+0=100\), less total bonuses paid \(15+35=50\), with \(L_0=L_3=0\). The negative third-year profit is not an error: it reflects the scheduled payment in a year with no pre-allocation result, partly offset by release of the previously established reserve.
Check. The reconciliation confirms that the declaration-year reserve increases and later release do not change total profit when the projection includes all payments and ends with no outstanding declared-bonus liability.
Interpretation and Common Exam Trap
Profit emergence is sensitive to both the timing of declarations and the amount allocated. A declaration can reduce current profit by creating or increasing a liability before cash is paid. A later payment reduces cash, but the release of its reserve offsets part of that reduction. This is why the year of payment cannot be analyzed in isolation from the reserve carried into that year.
The guarantee is preserved in these comparisons because \(S_t\) already reflects the contractual benefits and the discretionary amounts are treated separately. In an actual valuation, a proposed allocation must still comply with the policy terms and any applicable valuation requirements. The examples do not imply that an insurer may reduce a guaranteed benefit to offset a bonus.
- Do not confuse a declared bonus with its reserve. A future payment of \(B\) is valued as a liability at the projection date; it is not automatically a liability of \(B\) today.
- Do not subtract a bonus twice. If it is paid in year \(t\), include it in \(C_t\); if it remains unpaid at year-end, reflect it in \(L_t\). Apply the profit formula once to each effect.
- Do not infer total-profit invariance when the amount changes. With zero opening and ending reserves, changing total discretionary payments changes total profit by the same amount in the opposite direction.
- Do not compare annual profits without a common basis. The interest rate, payment date, reserve basis, and definition of \(S_t\) must be consistent across the alternatives.
The practical value of this analysis is that it separates two questions: how much surplus is allocated to policyholders, and when the accounting profit from the remaining surplus emerges. This treatment is aligned with the SOA Exam ALTAM (Advanced Long-Term Actuarial Mathematics) syllabus.
Check Your Understanding
For each problem, use the stated annual effective interest rate and treat all projections as deterministic, with no death or lapse during the period.
- A policy has \(S_1=45\). At the end of year 1, a bonus of \(12\) is declared, payable at the end of year 2. The annual effective interest rate is \(5\%\), and \(L_0=0\). Find year-1 profit. (Answer: \(33.57\))
- A two-year projection has \(S_1=40\), \(S_2=50\), and annual effective interest \(5\%\). A bonus of \(21\) is declared at the end of year 1 and paid at the end of year 2. Find both annual profits and total profit. (Answer: \(\pi_1=20.00,\ \pi_2=49.00,\ \text{total}=69.00\))
- A three-year projection has \(S_1=25\), \(S_2=55\), \(S_3=20\), with annual effective interest \(5\%\). Bonuses of \(10\) and \(20\) are declared at the ends of years 1 and 2, respectively, each payable one year later. Find the three annual profits and check total profit. (Answer: \(\pi_1=15.4762,\ \pi_2=35.4762,\ \pi_3=19.0476,\ \text{total}=70.0000\))
- Compare two allocations at annual effective interest \(4\%\), with \(S_1=30\), \(S_2=40\), \(S_3=30\). In the first, a bonus of \(24\) is declared and paid at the end of year 3. In the second, the same bonus is declared at the end of year 1 and paid at the end of year 3. Find the annual profits in each case and verify that total profit is unchanged. (Answer: First allocation \((30,40,6)\); second allocation approximately \((7.8107,39.1124,29.0769)\); total profit \(76.0000\) in each case)