Skip to content
Avrox Asset
Menu
The pension
Explore the tontine

The complete £100,000 example

Simulation results · September 2026

What income does the ladder produce?

Follow £100,000 through investments, survivor allocations and forty annual payment dates.

The experiment tests the worked ladder in pools of 50, 200 and 500 people, with different ages and balances. Each pool is run through 1,000 investment and mortality histories. The example member joins at 65 and remains the focus of the payment results. Each rung’s full proceeds are paid one year after release; the target sets the opening investments rather than fixing the payments.

Earlier payments are close to the funding target

At age 85, the simulated certainty equivalents are about £7,364, £7,437 and £7,446 for the three pools. Each is close to the ideal-sharing target of £7,420, and every eligible release at that age is completed.

The later rungs expose the harder question. In the smaller pool, some surviving members reach dates for which the policy has already stopped. The larger pools complete more of the calendar, but the oldest-age results depend on very few surviving observations.

Starting pool size

50 members · 1,000 simulated histories

Across these histories, the member reaches 20,936 payment release dates. The model completes 20,857 releases; 79 remain unresolved after a pool stops.

Payments supported by the same £100,000

FTP medianFTP 5th–95th percentileUnpooled medianIdeal sharing CE target
£0£6,250£12,500£18,750£25,00066758595105Age at payment
The band contains the middle 90% of annual payment outcomes. The solid lines show medians; the dotted line is the ideal-sharing CE target. Gaps mark incomplete coverage or fewer than 30 observations. All pool sizes use the same scale.
Can the simulated income support the CE target of £7,420?
Payment ageEligibleReleasedUnresolvedFTP CECE standard error
661,0001,0000£7,420£0
758618610£7,405£14
855805800£7,364£67
952031976
10058544
105936

CE values payments conditional on reaching release, using relative risk aversion 3. The standard error measures simulation sampling noise. A dash marks incomplete coverage or fewer than 30 observations; the downloads retain every outcome.

Where this pool reaches the current policy’s limits

886 of 1,000 pool histories (88.6%) stop before the final release over the full 39-year sharing horizon. This cumulative frequency describes the closed synthetic pool used in the experiment.

  • 596: fewer than three members remain with unreleased assets, the stated research stopping rule.
  • 290: the current state cannot meet the ordinary single-death FTP exposure condition.

A pool can stop after the focal member’s death or after many of their rungs have released. That is why pool continuation and individual payment coverage are shown separately. Unreleased assets remain recorded at the stop; their eventual settlement is an open contract question.

47,281 executed death events were audited. Maximum expected-transfer residual: £2.68e-9. The check uses floating-point arithmetic and a £0.0001 tolerance. Continuation is a separate test.

Follow one rung from its initial investment to payment

Recorded path 82, selected because the focal member reaches the final release age. This selection shows a long life rather than a representative history. The pool nevertheless stops at year 36.94; the later rungs remain recorded as retained assets.

Rung 20 · release at year 19, scheduled payment at year 20
Original investment£1,973.04
Return on that investment£2,473.87
Survivor allocations received£3,135.10
Return on survivor allocations£850.87
Released from longevity sharing£8,432.88
Return during the distribution year£300.38
Scheduled payment to member£8,733.25

After release, the entitlement belongs to the member and their beneficiaries. Download every released rung on this path (CSV).

200 members · 1,000 simulated histories

Across these histories, the member reaches 20,974 payment release dates. The model completes 20,972 releases; 2 remain unresolved after a pool stops.

Payments supported by the same £100,000

FTP medianFTP 5th–95th percentileUnpooled medianIdeal sharing CE target
£0£6,250£12,500£18,750£25,00066758595105Age at payment
The band contains the middle 90% of annual payment outcomes. The solid lines show medians; the dotted line is the ideal-sharing CE target. Gaps mark incomplete coverage or fewer than 30 observations. All pool sizes use the same scale.
Can the simulated income support the CE target of £7,420?
Payment ageEligibleReleasedUnresolvedFTP CECE standard error
661,0001,0000£7,420£0
758698690£7,432£10
855755750£7,437£55
951981980£7,421£162
10058580£6,511£418
10518162

CE values payments conditional on reaching release, using relative risk aversion 3. The standard error measures simulation sampling noise. A dash marks incomplete coverage or fewer than 30 observations; the downloads retain every outcome.

Where this pool reaches the current policy’s limits

153 of 1,000 pool histories (15.3%) stop before the final release over the full 39-year sharing horizon. This cumulative frequency describes the closed synthetic pool used in the experiment.

  • 128: the current state cannot meet the ordinary single-death FTP exposure condition.
  • 25: fewer than three members remain with unreleased assets, the stated research stopping rule.

A pool can stop after the focal member’s death or after many of their rungs have released. That is why pool continuation and individual payment coverage are shown separately. Unreleased assets remain recorded at the stop; their eventual settlement is an open contract question.

194,353 executed death events were audited. Maximum expected-transfer residual: £3.98e-10. The check uses floating-point arithmetic and a £0.0001 tolerance. Continuation is a separate test.

Follow one rung from its initial investment to payment

Recorded path 174, selected because the focal member reaches the final release age. This selection shows a long life rather than a representative history. The pool reaches the final scheduled release.

Rung 20 · release at year 19, scheduled payment at year 20
Original investment£1,973.04
Return on that investment£1,212.31
Survivor allocations received£1,946.84
Return on survivor allocations£446.71
Released from longevity sharing£5,578.90
Return during the distribution year£198.72
Scheduled payment to member£5,777.62

After release, the entitlement belongs to the member and their beneficiaries. Download every released rung on this path (CSV).

500 members · 1,000 simulated histories

Across these histories, the member reaches 20,585 payment release dates. The model completes 20,585 releases; 0 remain unresolved after a pool stops.

Payments supported by the same £100,000

FTP medianFTP 5th–95th percentileUnpooled medianIdeal sharing CE target
£0£6,250£12,500£18,750£25,00066758595105Age at payment
The band contains the middle 90% of annual payment outcomes. The solid lines show medians; the dotted line is the ideal-sharing CE target. Gaps mark incomplete coverage or fewer than 30 observations. All pool sizes use the same scale.
Can the simulated income support the CE target of £7,420?
Payment ageEligibleReleasedUnresolvedFTP CECE standard error
661,0001,0000£7,420£0
758638630£7,430£8
855485480£7,446£55
951901900£7,332£184
10052520£6,367£524
105770

CE values payments conditional on reaching release, using relative risk aversion 3. The standard error measures simulation sampling noise. A dash marks incomplete coverage or fewer than 30 observations; the downloads retain every outcome.

Where this pool reaches the current policy’s limits

5 of 1,000 pool histories (0.5%) stop before the final release over the full 39-year sharing horizon. This cumulative frequency describes the closed synthetic pool used in the experiment.

  • 5: the current state cannot meet the ordinary single-death FTP exposure condition.

A pool can stop after the focal member’s death or after many of their rungs have released. That is why pool continuation and individual payment coverage are shown separately. Unreleased assets remain recorded at the stop; their eventual settlement is an open contract question.

486,666 executed death events were audited. Maximum expected-transfer residual: £1.02e-10. The check uses floating-point arithmetic and a £0.0001 tolerance. Continuation is a separate test.

Follow one rung from its initial investment to payment

Recorded path 13, selected because the focal member reaches the final release age. This selection shows a long life rather than a representative history. The pool reaches the final scheduled release.

Rung 20 · release at year 19, scheduled payment at year 20
Original investment£1,973.04
Return on that investment£2,230.43
Survivor allocations received£2,920.31
Return on survivor allocations£662.71
Released from longevity sharing£7,786.50
Return during the distribution year£277.35
Scheduled payment to member£8,063.85

After release, the entitlement belongs to the member and their beneficiaries. Download every released rung on this path (CSV).

Eligibility is assessed at release, one year before payment. An entitlement released before a member’s death is paid to their beneficiary. Each pool size uses a different fixed cohort, so differences reflect composition as well as size.

The targets being tested

Annual targets from the same £100,000 contribution

£4,907Annual investment-only CE target.
Independently funded unpooled ladder.
£7,420Annual ideal-sharing CE target.
The finite-pool simulation tests delivery.
Same contribution, strategy and forty payment dates; different initial rung weights. The ideal-sharing model raises the level target by 51.2%. Actual finite-pool payments are examined in the simulations.

The two ladders each begin with £100,000 and use the same investment strategy and monthly market shocks. Each is independently allocated to fund its own level target. The unpooled assets remain available to the member or their estate; the pooled ladder shares unreleased balances.

The funding schedule shows every opening investment. The comparisons here ask how actual payments measure up to those starting targets.

Assumptions

The model uses synthetic independent mortality and a specified equity-to-bond glidepath. All amounts are nominal. Relative risk aversion is 3. The policy follows known death order and checks each ordinary single-death FTP allocation.

People, investments and dates
Members
The example member contributes £100,000 at age 65. Peers’ ages span 55–80 and contributions £60,000–£140,000, independently shuffled. All use the same forty calendar payment dates. There are no later entrants or contributions.
Investment policy
Equities have a 6.5% annual GBM drift and 16% volatility. Their weight falls from 80% at thirty years before release to zero at five years. Bonds earn a deterministic continuously compounded 3.5%; the distribution account earns the same rate for its final year.
Mortality
The monthly hazard is 0.01 × exp((entry age − 65 + elapsed years) / 10), held constant within each month. The same law supplies survival probabilities and death times.
Scope
The experiment excludes fees, tax, inflation, mortality improvements, correlated deaths and external buffers. The horizon ends with payment at age 105 for the example member.
Allocation, research stops and accounting

Survivor receipts are allocated across the recipient’s remaining rungs using conditional-survival and remaining CE weights. Existing investments stay in their own rungs.

The calculation stops when the current single-death exposure is infeasible, an audit fails, or fewer than three members retain unreleased assets. Those assets remain in the ledger pending a settlement policy. Already released entitlements continue to their scheduled payment dates.

Recorded payments include the eventual cash flows from already released assets. Retained assets are valued at the stopping date. The model note records this accounting convention and the full equations.

The next question is continuation

The experiment connects an opening funding rule to the income produced by actual FTP transfers. It also identifies the paths on which the current policy stops, leaving future payment outcomes unresolved.

The next study needs a continuation or transfer rule that accounts for those retained entitlements. The resulting complete payout distribution can then be used to recalibrate the starting investments. The capacity research supplies candidate constraints for that work.

Source and results

The package contains the generator, assumptions, fixed cohorts, every example-member outcome, pool stop records and selected event ledgers. Independent checks reconcile the funding, allocations and payment accounts.