Actuarial Mathematics for Modelling · Projecting and valuing cashflows contingent on multiple decrement events
Projecting Cashflows with Multiple Decrements
Updated 11 October 2026 · Fact-checked
Projecting cashflows with multiple decrements means tracking how many lives stay in force each year, then applying each cause's decrement rate to find expected exits. Multiply exits by the benefit for that cause, discount from the payment date, and sum. Survivors at each date are the starting group for the next year.
Understand Projecting Cashflows with Multiple Decrements
A decrement is any event that removes a life from a group or a policy from the in-force book. Death, withdrawal, ill-health and retirement are common examples. Each cause can trigger a different cashflow, so you cannot treat all exits alike.
Start with a group of lives at a given age. Over one year, some leave by death, some by withdrawal, and so on. The rest stay. The next year starts with only those who stayed. This is why a projection is done year by year, in order.
Each cause has a decrement rate. In the IAI notation, (aq)^j_x is the probability that a life in service at age x leaves in the year by cause j, with all causes acting together. Because all causes act together, the exit probabilities for one year add up. The survival probability is one minus that total.
An expected cashflow is the number of lives expected to exit by a cause, times the amount paid for that cause. For a single policy, the 'number' becomes a probability. Premiums and other in-force payments depend on being in force at that date. Exit benefits depend on leaving by a given cause in a given year.
Once you have expected cashflows at each time, you can discount them to get an expected present value. This is the building block for valuing benefits contingent on decrement events and for profit tests.
Key rules to remember
- Number in force after one year
- (al)_{x+1} = (al)_x − Σ_j (ad)^j_x
- Sum over all causes j. Use the same year's decrements for all causes.
- Decrements by cause
- (ad)^j_x = (al)_x × (aq)^j_x
- Use the dependent (all causes acting) rate (aq)^j_x. This gives the expected number leaving by cause j between age x and x+1.
- Rate from table
- (aq)^j_x = (ad)^j_x ÷ (al)_x
- Use this when the question gives a multiple decrement table instead of rates.
- Probability of staying in force one year
- (ap)^τ_x = 1 − Σ_j (aq)^j_x = (al)_{x+1} ÷ (al)_x
- τ means all causes combined.
- Probability in force t years ahead
- t(ap)^τ_x = (al)_{x+t} ÷ (al)_x = (ap)^τ_x × (ap)^τ_{x+1} × … × (ap)^τ_{x+t−1}
- Multiply the one-year survival probabilities of consecutive ages.
- Expected cashflow for cause j in year t+1 (single policy)
- t(ap)^τ_x × (aq)^j_{x+t} × B_j
- B_j is the benefit payable on exit by cause j. Payment is at the time stated, for example the end of the year of exit.
- Expected present value
- EPV = Σ_t E(CF_t) × v^t
- Use v = 1 ÷ (1 + i). If exits happen mid-year, discount from the actual payment time, using the stated timing.
How to solve Projecting Cashflows with Multiple Decrements questions
Use this order for any question that asks you to project expected cashflows with several exit causes.
- 1Write down the starting group size, or probability 1 for a single policy, and the age or time at which you start.
- 2List every cause of decrement and the rate for each, for each age. Note which rates are dependent (aq)^j_x. If the question gives independent rates, convert first as the question requires.
- 3For each year, compute exits by cause: starting number × (aq)^j_x. Subtract all of them to get the starting number for the next year.
- 4Attach the benefit to each cause. Check the payment timing: start of year, end of year, or on exit. Also work out in-force cashflows such as premiums or salaries, based on the number in force at the right date.
- 5Form the expected cashflow at each time point by adding all inflows and outflows due then. Keep the sign convention clear.
- 6If asked, discount each cashflow with v^t from the payment date and add them to get the EPV.
- 7Check: exits plus survivors equal the starting number each year, and survivors never go up.
Quickest way: Year-by-year table
When to use it: Use this for almost every numerical question. It is fast, shows method for marks, and makes errors easy to spot.
- Draw a table with columns: year, in force at start, one column for exits by each cause, in force at end.
- Fill the first row from the given rates. Carry the end number to the next row's start.
- Add a column for expected cashflow: exits × benefits, plus in-force items.
- Add a column for v^t and a column for PV only if discounting is asked. Total the PV column last.
- Do a quick check on each row: start − exits = end.
Common mistakes in Projecting Cashflows with Multiple Decrements
Applying the rate to the original group every year instead of the survivors.
It feels simpler to multiply the initial number by each year's rate.
Fix: Always use the number in force at the start of that year. Update it every row.
Paying a benefit for every life who leaves, ignoring which cause it was.
Students total all exits and use one benefit amount.
Fix: Compute exits separately by cause and multiply each by its own benefit.
Discounting from the wrong time, for example discounting an end-of-year benefit by v^0.
Mixing up the time of exit with the time the payment is made.
Fix: Write the payment time next to each cashflow before discounting. Use the timing given in the question.
Charging a premium or other in-force payment to those who just left.
Students forget that in-force cashflows depend on survival to that date.
Fix: For a start-of-year premium in year t+1, use the number in force at time t, after earlier exits and before this year's exits.
Mixing dependent and independent decrement rates.
The notation looks similar, and the question may give one type but the table needs the other.
Fix: Check whether the given rates are the all-causes-acting (aq)^j_x. Only these can be summed to give the exit probability for the year.
Rounding expected numbers of lives to whole numbers during the projection.
Students think half a person cannot exit.
Fix: Keep fractions throughout. These are expected values. Round only the final answer.
Worked examples
Example 1
A scheme has 1,000 members aged 62 at the start of the year. Decrement rates (all causes acting) at age 62 are: death 0.010, withdrawal 0.040, retirement 0.150. At age 63 they are: death 0.012, withdrawal 0.030, retirement 0.200. Benefits are paid at the end of the year of exit: death ₹5,00,000, withdrawal ₹1,00,000, retirement ₹8,00,000. Find the expected benefit outgo at the end of each of the first two years and its present value at 5% per year.
Show the solution
- Year 1 exits: deaths 1,000 × 0.010 = 10; withdrawals 1,000 × 0.040 = 40; retirements 1,000 × 0.150 = 150.
- Survivors at age 63: 1,000 − 10 − 40 − 150 = 800.
- Year 1 outgo: 10 × ₹5,00,000 = ₹50,00,000; 40 × ₹1,00,000 = ₹40,00,000; 150 × ₹8,00,000 = ₹12,00,00,000. Total ₹12,90,00,000.
- Year 2 exits from 800: deaths 800 × 0.012 = 9.6; withdrawals 800 × 0.030 = 24; retirements 800 × 0.200 = 160.
- Survivors at age 64: 800 − 9.6 − 24 − 160 = 606.4.
- Year 2 outgo: 9.6 × ₹5,00,000 = ₹48,00,000; 24 × ₹1,00,000 = ₹24,00,000; 160 × ₹8,00,000 = ₹12,80,00,000. Total ₹13,52,00,000.
- PV: 12,90,00,000 ÷ 1.05 = ₹12,28,57,143 (approx). 13,52,00,000 ÷ 1.05² = ₹12,26,30,385 (approx).
- Sum: ₹24,54,87,528 (approx).
Answer: Expected outgo is ₹12,90,00,000 at the end of year 1 and ₹13,52,00,000 at the end of year 2. The present value at 5% is about ₹24,54,87,528.
Example 2
A policy is issued to a life aged 50. Annual premium of ₹20,000 is paid at the start of each year while the policy is in force. On death in a year, ₹10,00,000 is paid at the end of that year. On withdrawal in a year, a surrender value is paid at the end of that year: ₹30,000 in year 1 and ₹60,000 in year 2. Decrement rates (all causes acting): year 1 death 0.002, withdrawal 0.10; year 2 death 0.0025, withdrawal 0.08. The policy has a term of 2 years, so no premium is paid at time 2 and no benefit is paid after time 2. Find the expected net cashflow (premiums less benefits) at times 0, 1 and 2, and the EPV at 5% per year.
Show the solution
- Probability in force at time 0 = 1. Probability in force at time 1 = 1 − 0.002 − 0.10 = 0.898.
- Time 0: premium ₹20,000 × 1 = ₹20,000. No benefits yet. Net = +₹20,000.
- Time 1 premium: 0.898 × ₹20,000 = ₹17,960.
- Time 1 death benefit: 0.002 × ₹10,00,000 = ₹2,000. Time 1 withdrawal: 0.10 × ₹30,000 = ₹3,000. Benefits = ₹5,000.
- Time 1 net: 17,960 − 5,000 = +₹12,960.
- Time 2 death benefit: 0.898 × 0.0025 × ₹10,00,000 = ₹2,245.
- Time 2 withdrawal: 0.898 × 0.08 × ₹60,000 = 0.07184 × 60,000 = ₹4,310.40.
- Time 2 net: −(2,245 + 4,310.40) = −₹6,555.40. There is no premium at time 2 because the policy has a term of 2 years, as stated in the question.
- EPV = 20,000 + 12,960 ÷ 1.05 − 6,555.40 ÷ 1.05² = 20,000 + 12,342.86 − 5,945.94 = ₹26,396.92 (approx).
Answer: Expected net cashflows are +₹20,000 at time 0, +₹12,960 at time 1 and −₹6,555.40 at time 2. The EPV at 5% is about ₹26,397.
Exam tips
- Draw the table even when the question seems short. Examiners award marks for method, and the table lets them follow your working.
- Read the payment timing twice: start of year, end of year, or at the moment of exit. This changes both the in-force number used and the discount power.
- State your assumptions in a line, such as 'exits occur at year end' or 'rates are all-causes-acting'. Written questions reward clear assumptions.
- In Paper B (computer-based), build the projection in Excel with one row per year and rates in separate input cells. Link formulas so changing a rate updates everything.
- Check that exits plus survivors equal the starting number in each row. This catches most arithmetic slips in a few seconds.
Practice questions from Projecting and valuing cashflows contingent on multiple decrement events
- A policy pays Rs 200,000 immediately on death within 10 years. The only other decrement is withdrawal, which pays nothing. Forces are consta…
- For a scheme member, the dependent probabilities of leaving service by any cause are 0.08 in year 1 and 0.10 in year 2. In year 3, the depen…
- An employee aged 55 is subject to two decrements with constant forces: death 0.02 and withdrawal 0.08 per year. What is the probability that…
- Which statement about dependent and independent decrement rates is correct for a life subject to two causes of decrement?
- A population of 1,000 lives is subject to two decrements, death and withdrawal, with constant forces of 0.01 and 0.04 per annum respectively…
Projecting Cashflows with Multiple Decrements: frequently asked questions
What is the difference between a decrement and a survivor in a projection?
A decrement is a life that leaves the group by one of the listed causes during the year. A survivor stays in force at the end of the year. Survivors form the starting group for the next year.
Do I need to discount when projecting expected cashflows?
Only if the question asks for a present value or EPV. A projection alone gives expected amounts at each time. Discounting uses v^t from the payment time to time 0.
How do I handle a decrement table instead of rates?
Divide each decrement count (ad)^j_x by the number in force (al)_x to get the rate, or work directly with the counts scaled to your group size. Use ratios such as (al)_{x+t} ÷ (al)_x for survival to later ages.
What if exits happen during the year rather than at the end?
Use the timing given in the question. If benefits are paid at the moment of exit, discount from the exit time, often assumed mid-year if the question says so. State your assumption clearly.