Why Profit Timing Matters
A product can have positive projected profits over its lifetime and still require substantial capital during some durations. The reason is timing: expenses, claims, reserve strengthening, guarantee costs, or declared bonuses may reduce profit early, while expected gains may emerge later. A single lifetime total hides this pattern.
A profit signature makes the timing visible. Read alongside projected reserves and required capital, it helps identify when financial pressure may be greatest. It is evidence for investigation, not by itself a solvency measure or proof of why strain occurs. The discussion and examples below use hypothetical projections and are aligned with the SOA Exam ALTAM (Advanced Long-Term Actuarial Mathematics) syllabus.
In an actuarial projection, the profit amounts may reflect premiums, expenses, claims, investment income, and changes in reserves under a specified basis. A profit amount is meaningful only with its basis, unit, and timing stated. For example, a negative year-end profit could reflect a high benefit outgo, a reserve increase, or both. The signature alone does not identify which component caused it.
Discounting a Profit Signature
Suppose each \(P_t\) is paid or recognized at the end of year \(t\), and the effective annual interest rate is \(i\). Let \(v=1/(1+i)\). The discounted profit total at time 0 is the sum of each profit discounted from its own payment time:
This is a net present value of the projected profit stream on the stated basis. It is not a projection of required capital, and it does not indicate whether capital remains adequate at every intermediate time.
Worked Example: Discounting a Signature with an Early Loss
A hypothetical three-year portfolio has projected year-end profits of \(-\$12{,}000\), \(\$7{,}000\), and \(\$10{,}000\). Find the time-0 discounted profit total at an effective annual interest rate of \(5\%\), and describe what the signature shows. The supplied profits are aggregate projection amounts; assume the projection already incorporates its stated mortality assumption of \(1.0\%\) per year and lapse assumption of \(4.0\%\) per year. No separate decrement calculation is required.
Method. Discount each year-end profit by the factor for its duration, then add the unrounded present values. This avoids treating a later dollar as if it were available at time 0.
The discount factors for years 1, 2, and 3 are:
The year-1 present value is \(-12{,}000/1.05=-\$11{,}428.57143\). The year-2 present value is \(7{,}000/1.05^2=\$6{,}349.20635\). The year-3 present value is \(10{,}000/1.05^3=\$8{,}638.37599\). Therefore:
The signature begins with a substantial loss and then has two positive years; its discounted total is positive. The result is rounded once, after summing the unrounded values. Adding the displayed amounts rounded to cents would give a slightly different sum, so that is not the method used here. The positive NPV does not show whether capital is sufficient during year 1.
Compare Profit with Available and Required Capital
A solvency projection supplies a second time profile. Let \(A_t\) be available capital at time \(t\), and \(K_t\) be required capital at that time. Define the solvency margin as \(M_t=A_t-K_t\). A negative margin means available capital is below the stated requirement at that projection time; a smaller margin indicates less headroom than a larger one.
The profit signature and capital schedule answer related but different questions. The signature shows when projected profits arise. The margin schedule compares available capital with required capital at each time. Required capital can change independently of profits, for example as the projected risk exposure or guarantee burden changes. Consequently, neither a positive NPV nor a positive final margin guarantees a nonnegative margin in every intervening year.
Worked Example: Finding the Duration of Greatest Capital Strain
A hypothetical portfolio starts with \(\$30{,}000\) of available capital. Its year-end profits are \(-\$6{,}000\), \(\$4{,}000\), and \(\$9{,}000\). Required capital at times 0, 1, 2, and 3 is \(\$22{,}000\), \(\$28{,}000\), \(\$29{,}000\), and \(\$24{,}000\), respectively. Find available capital and the solvency margin at each time, then identify the greatest strain. Profits are supplied aggregate amounts on a hypothetical projection basis with annual mortality of \(1.0\%\) and lapse of \(4.0\%\) already reflected. No interest calculation is needed for this capital roll-forward.
Method. Roll available capital forward in chronological order, adding each end-year profit, and subtract the required capital at the matching time. This prevents a year-end profit from being used to support capital at an earlier time.
At time 0, \(A_0=\$30{,}000\), so \(M_0=30{,}000-22{,}000=\$8{,}000\).
At time 1, \(A_1=30{,}000-6{,}000=\$24{,}000\), so \(M_1=24{,}000-28{,}000=-\$4{,}000\).
At time 2, \(A_2=24{,}000+4{,}000=\$28{,}000\), so \(M_2=28{,}000-29{,}000=-\$1{,}000\).
At time 3, \(A_3=28{,}000+9{,}000=\$37{,}000\), so \(M_3=37{,}000-24{,}000=\$13{,}000\).
The margins are therefore \(\$8{,}000\), \(-\$4{,}000\), \(-\$1{,}000\), and \(\$13{,}000\). The lowest margin is at time 1, so that is the duration of greatest strain on this measure. The time-3 profit cannot be counted in available capital at time 1.
Same NPV, Different Capital Pressure
A useful interpretation test is to compare products with similar discounted totals but different profit timing. The comparison is informative only when valuation time, interest rate, profit units, and projection horizon are consistent. Even then, capital requirements may differ, so a profit comparison alone is not a solvency comparison.
Worked Example: Equal Discounted Profits but Unequal Interim Margins
Two hypothetical two-year product designs have year-end profits, in thousands of dollars, as follows: Design A has \((-10,12)\), while Design B has \((-4,5.7)\). Use an effective annual interest rate of \(5\%\). Each starts with \(\$15{,}000\) of available capital. Required capital, in thousands, is 10 at time 0, 12 for A and 8 for B at time 1, and 14 for each at time 2. Assume annual mortality of \(1.0\%\) and lapse of \(4.0\%\) have already been reflected in the supplied profit projections, and assume no other capital movements. Compare the NPVs and margins.
Method. First calculate each NPV using year-specific discount factors. Then independently roll forward capital and compare it with each design's required capital. This avoids assuming that equal NPVs imply equal capital profiles.
For Design A, the year-1 present value is \(-10/1.05=-9.52380952\) thousand, and the year-2 present value is \(12/1.1025=10.88435374\) thousand. Thus its NPV is \(1.36054422\) thousand.
For Design B, the year-1 present value is \(-4/1.05=-3.80952381\) thousand, and the year-2 present value is \(5.7/1.1025=5.17006803\) thousand. Its NPV is also \(1.36054422\) thousand. Both NPVs are approximately \(\$1{,}360.54\).
At time 0, each design has available capital of \(\$15{,}000\) and required capital of \(\$10{,}000\), giving a margin of \(\$5{,}000\). At time 1, A has \(15-10=5\) thousand available, so its margin is \(5-12=-7\) thousand. B has \(15-4=11\) thousand available, so its margin is \(11-8=3\) thousand.
At time 2, A has \(5+12=17\) thousand available and a margin of \(17-14=3\) thousand. B has \(11+5.7=16.7\) thousand available and a margin of \(16.7-14=2.7\) thousand. Although the discounted profit totals match, A has much greater interim strain in this illustration. The example shows why a total alone cannot replace a duration-by-duration capital comparison.
Interpreting Reserves, Guarantees, and Bonus Commitments
Reserve projections help explain the profit pattern, but reserve levels and required capital are not interchangeable. A reserve is a liability measure on a specified valuation basis; required capital is a separate measure of resources needed under the applicable solvency framework. A reserve increase can reduce current reported profit under a given profit definition, while a later release may contribute to later profit. Neither effect automatically determines the required-capital path.
For a hypothetical product with a guarantee, inspect whether the projected periods of weak profit coincide with projected guarantee costs, benefit outgo, or reserve increases. For a participating product, inspect the timing of bonus commitments and their associated cash flows or reserve effects. Then compare those observations with available and required capital at the same durations. A coincidence is a reason to investigate the product mechanics and assumptions; it is not enough, by itself, to establish causation.
For a disciplined review, keep the following quantities distinct:
| Quantity | What it summarizes | What it does not establish by itself |
|---|---|---|
| Profit signature | Projected profit by duration | Whether capital meets requirements at every time |
| Discounted profit total | Present value of profits on a stated basis | The timing or severity of interim capital strain |
| Reserve projection | Projected liability values on a specified basis | The solvency margin without available and required capital |
| Solvency margin | Available capital less required capital at a stated time | The particular profit component that caused a low margin |
Common Exam Trap: Mixing Timing and Measures
A frequent error is to report a positive discounted profit total as proof that a product never experiences capital strain. That conclusion does not follow: the NPV aggregates profits after discounting, whereas a margin is evaluated at a particular time and depends on both available and required capital.
For example, using Design A above, an incorrect check might compare its initial \(\$15{,}000\) of capital only with its positive \(\$1{,}360.54\) NPV and conclude that there is no strain. That is not a solvency-margin calculation. At time 1, available capital is only \(\$5{,}000\), while required capital is \(\$12{,}000\), so the actual stated margin is \(-\$7{,}000\). The correction is to project capital in time order and compare \(A_t\) with \(K_t\) at each matching time.
A second timing error is discounting every profit by \(v\), regardless of duration. In the three-year example, the year-3 profit must be discounted by \(v^3\), not \(v\). Match each profit to its own time before summing. Round only after calculating a total from unrounded values.
Check Your Understanding
For each problem, keep profit timing separate from the capital comparison. Monetary amounts are in thousands unless otherwise stated.
- A hypothetical two-year portfolio has year-end profits of \(-2\) and \(5\), and an effective annual interest rate of \(10\%\). Find the time-0 discounted profit total. (Answer: \(2.31\) thousand.)
- A portfolio starts with available capital of \(20\). Its profits are \(-4\) in year 1 and \(7\) in year 2. Required capital at times 0, 1, and 2 is \(15,18,25\). Assume no other capital movements. Find all three margins. (Answer: \(5,-2,-2\) thousand.)
- At an effective annual interest rate of \(4\%\), a product has year-end profits \((-5,4,4)\). Find the discounted profit total to the nearest \(0.01\) thousand. (Answer: \(2.45\) thousand.)
- A product has a positive discounted profit total, but its projected available capital at time 1 is \(16\) and required capital is \(19\). State the time-1 margin and explain whether the positive total changes its interpretation. (Answer: The margin is \(-3\); the positive total does not remove the time-1 shortfall.)
- Two designs have identical NPVs. Design X has margins \(6,-4,5\) thousand at times 0, 1, and 2; Design Y has margins \(6,2,5\) thousand. Which has greater interim strain on the stated margin measure, and at what time? (Answer: Design X, at time 1, with a margin of \(-4\) thousand.)