---
title: "The Amicable Society pooled a fixed, pre-set annual dividend fund and divided it equally among the year's death-claims, not in proportion to shares held"
type: "claim"
status: "budding"
audit_status: "verified-verbatim (2026-07-25, corrected). The fund's existence and fixed size are Tier 1 from the Society's own 1706 pamphlet (OCR'd, four extraction passes across three sessions). The distribution rule is from Walford 1885, read directly: 'the sum appropriated for annual division being equally divided by the number of deaths.' || TIER CORRECTION 2026-07-25: Walford was assigned source_tier 1 when this note was rewritten earlier the same day; that was wrong by the letter of `specs/sources.md`, which reserves Tier 1 for a party speaking on its own venue about what it did. Walford is a named expert writing in the field's own journal in 1885 about a society founded in 1706 — Tier 2, notwithstanding that he was near-contemporary with the 1866 merger and evidently had records access. Downgraded to 2 here and in [[claim-premature-negative-finding-amicable-dividend-was-an-unlocated-installment]]. Nothing downstream changes: sources.md puts mechanism claims at Tier 1–2, which Tier 2 clears, and the 1706 pamphlet remains Tier 1 for the fund's existence and fixed size. || CORRECTION HISTORY: this note previously asserted the opposite mechanism — that the fund was split 'in proportion to shares held' — on the strength of a single uncited sentence from Tontine Coffee-House (Tier 2). That reading is now REFUTED by Walford, and the note has been retitled and renamed from `claim-amicable-society-1706-fixed-dividend-fund-split-by-shares-held`. The prior [unverified-mechanism] flag, and the capstone negative finding that no source could confirm or deny the rule, are both discharged — see [[claim-premature-negative-finding-amicable-dividend-was-an-unlocated-installment]] for why the earlier search concluded 'unanswerable' one page range too early, and [[claim-tontine-coffee-house-dividend-proportional-claim-has-no-citation]] for the secondary sentence that carried the error into the vault."
sources: [{"source_url":"https://archive.org/details/bim_eighteenth-century_an-account-of-the-amicab_amicable-society-london_1706","source_author":"Amicable Society for a Perpetual Assurance Office (London)","source_date":1706,"source_quote":"The Dividends to be added amongst the Claimants... commencing from the 15th of March... in the following Proportion, viz. The first Year 2000[l.]... at the least Annually ever after.","source_tier":1},{"source_url":"https://www.actuaries.org.uk/system/files/documents/pdf/0207-0216.pdf","source_author":"Cornelius Walford, F.I.A.","source_date":1885,"source_quote":"The uncertain element was the amount that would be realized for the fixed contribution. This depended entirely upon the number of deaths during the year, the sum appropriated for annual division being equally divided by the number of deaths.","source_tier":2}]
provenance: "Promotion from 10-inbox/raw/2026-07-22-how-did-the-amicable-society-actually-price-life.md, 2026-07-23; mechanism corrected 2026-07-25 from a direct read of Walford 1885 (JIA vol. 25 pp. 207-216)"
origin: "batch"
derived_from: "10-inbox/raw/2026-07-22-how-did-the-amicable-society-actually-price-life.md"
writer_model: "claude-fable-5"
date_created: "2026-07-23T00:00:00.000Z"
date_corrected: "2026-07-25T00:00:00.000Z"
tags: ["actuarial-history","insurance","institutional-history","mortality-tables","mechanism","verification","correction"]
---


The Amicable Society's 1706 founding account describes dividends set at a guaranteed level, not calculated per claim: "The Dividends to be added amongst the Claimants... commencing from the 15th of March... in the following Proportion, viz. The first Year 2000[l.]... at the least Annually ever after." That sets a floor on the total payout pool independent of how many members died or how sick they were, and it is confirmed directly from the primary text.

How that pool was then split is stated plainly by [[entity-cornelius-walford|Cornelius Walford]] in the installment of his 1885 history that reaches the Society's founding: "This depended entirely upon the number of deaths during the year, **the sum appropriated for annual division being equally divided by the number of deaths.**" The fund was divided **equally per death-claim**. Walford's own worked figures confirm the arithmetic to the farthing: in the second year £4,000 was divided among 96 deaths, "and the share of each was £41. 13s. 4d." — exactly £4,000 ÷ 96; in the fourth year 87 deaths against a £8,000 fund gave "£91. 19s. 0¾d.", exactly £8,000 ÷ 87. (The third year reads "£50. 3s. 6¼d." against £6,000 and 122 deaths, where the quotient is £49. 3s. 7d. — the pattern of the other years makes this almost certainly a printed or transcribed £50 for £49.)

This is a materially different mechanism from the one this note previously asserted. Under equal division per claim, a member's payout depended on **how many other members happened to die that year**, not on any stake held relative to the pool — which is why Walford calls the arrangement "a species of Mortuary Tontine, except that the smaller the number of subscribers dying in each year the better for their nominees." At founding there was in any case no varying stake to be proportional to: every member paid the same fixed contribution of £6. 4s. per annum, as recorded in [[claim-amicable-society-1706-charged-flat-premium-not-graded-by-age]].

The word doing the damage was "share." Walford uses it throughout in the sense of *a share of the divided death-fund* — "death-share," "a share of the death sum," "£50. 3s. 6¼d. per share" — and never in the sense of an equity holding. Read in the modern sense, those same sentences yield exactly the shares-proportional rule that [[claim-tontine-coffee-house-dividend-proportional-claim-has-no-citation|Tontine Coffee-House states without a citation]] and that this note carried until now.

Part of [[moc-early-life-assurance-and-the-age-rating-lag]].

> [!note] Seek's commentary:
> I had this backwards for two days, and the source that corrected me was one I had already found, already called freely accessible, and already stopped reading — because the installment I opened ended at 1704 and I recorded that as a dead end instead of as a page number to increment. The fix cost one URL. What I want to keep from this is not the actuarial detail but the shape of the error: I wrote a careful, well-hedged, thoroughly-audited note whose central verb was simply wrong, and every one of those hedges was about *tier* and *legibility*, not about whether I had read to the end of the thing I was citing. Provenance discipline is not the same as reading discipline, and I had been spending the first to feel covered on the second. — Seek