Early English life assurance — the seventy-year lag between age-rated pricing and anyone using it

The vault accumulated this cluster sideways — it began as a question about how one 1706 mutual priced its policies — and what it turned into is a case study in institutional lag. The mathematics of age-rated life pricing was published in 1693. The first institution to actually price that way opened in 1762. The society at the centre of this cluster did not adopt age-graded premiums until 1807, a hundred and one years after its own founding. Nothing was secret, nothing was lost, and nobody was waiting on a discovery. The method sat in the Philosophical Transactions for three generations while the institutions that most needed it went on charging everyone the same.

The spine: method in 1693, adoption in 1762

• claim-halley-1693-breslau-table-founded-annuity-pricing — Halley builds the first life table from real population data (Breslau parish registers), founding the pricing of annuities on lives.

• claim-halley-1693-age-scaled-annuity-prices-britain-kept-flat — the load-bearing one. Halley did not stop at the table; he computed annuity values as a function of the nominee's age, discounting at 6%. The British government kept selling annuities at a single flat price for decades anyway. The lag starts here, in the same paper that closes the technical question.

• claim-amicable-society-1706-charged-flat-premium-not-graded-by-age — thirteen years later the Amicable Society opens with one fee schedule for "every Subscriber." Age is not an input.

• claim-amicable-society-1706-rules-capped-admission-age-at-55 — age enters only as a pass/fail gate at the door: a coarse eligibility band, not a rating variable. The band tightened over time: 12–55 at the founding per the Society's own 1706 pamphlet, reduced to 45 by around 1770 per Walford. (Walford himself places the explicit 12–55 limit only "at a later period" and says there was "no limitation of age for membership in the first instance" — a primary-vs-secondary discrepancy the linked note records rather than resolves.)

• claim-dodson-refused-amicable-society-over-age-45 — the gate turns out to be the hinge of the whole story. James Dodson is refused admission in the 1750s for exceeding the cap, under the tightened 45 rather than the original 55.

• claim-dodson-built-age-scaled-premiums-founding-equitable-life — and within months designs the thing the Amicable's own rules made necessary: premiums correctly scaled to age, advertised at a meeting on 2 March 1756 and set out in a lecture that year, becoming the Society for Equitable Assurances in 1762. Dodson died in 1757 and never saw it. Note the joint carefully — then, not therefore. See claim-dodson-rejection-causation-is-reported-tradition-not-documented.

• claim-amicable-society-adopted-age-graded-premiums-only-in-1807 — and the lag is longer than 1762, because the incumbent did not follow. The Amicable charged a flat contribution for another forty-five years after its competitor demonstrated the alternative, adopting "a graduated scale of Contributions, according to age and circumstance" only under its 1807 charter — 101 years after its own founding, 114 after Halley.

Read end to end, the sequence is not "the maths wasn't ready." Halley's paper contains the method, and the institution that eventually used it was founded on the design of a man the incumbent had turned away on the basis of the very variable it refused to price. (Whether the turning-away produced the design is the one joint in this chain the sources do not evidence — see the open-threads note below.) Nor is it "the method existed and then someone adopted it": a working demonstration by a direct competitor was also insufficient, and what finally changed the incumbent's pricing was a rewritten charter.

What the Amicable actually did instead

Because it did not price by age, the Society needed some other way to make the numbers work. That mechanism is now documented directly from Walford 1885, and it is stranger than a simple flat premium:

• claim-amicable-society-1706-fixed-dividend-fund-divided-equally-per-death — a fixed, pre-set annual fund, divided equally among that year's death-claims. Your family's payout depended on how many other members died that year, not on anything about you. Walford: "a species of Mortuary Tontine, except that the smaller the number of subscribers dying in each year the better for their nominees."

So risk was not priced, it was pooled and rationed: the Society fixed what it would pay out in total and let the claimants divide it. Age screening at the door was the only underwriting in the system, which is exactly why the door was where Dodson hit it.

Two consequences of that design, both from the same Walford installment:

• claim-1711-amicable-policy-sale-possibly-first-life-policy-sale-on-record — because a policy's payout depended on how many other members died in a given year, and the realised figures swung from £30 to £92 per claim in the Society's first four years, holders had a reason to sell certainty and buyers had a reason to bet on mortality. Policies were offered for sale by 1711, and the Society was keeping a register of assigned policies by 1733.
• claim-amicable-registrar-hartley-absconded-with-6500-in-1713 — the design also required a large undivided surplus to accumulate, and in 1713 the founder-registrar left with £6,500 of it. A rival society had published a warning naming that exact temptation before it happened.

Afterwards: the table that replaced Halley's

• claim-price-northampton-table-standard-actuarial-reference-century — Richard Price's Northampton table and Observations on Reversionary Payments (1771) become the standard reference for roughly a century.
• claim-price-northampton-table-overestimated-mortality-favouring-insurers — and it was biased: it overestimated mortality, which is good for the insurer selling assurance and bad for the annuitant. The standard reference of the age-rating era was systematically wrong in a direction that favoured the institutions using it.
• claim-pensam-1811-amicable-experience-first-assured-lives-mortality-data — and the first check on it came from the least actuarial institution in the story. Around 1811 the Amicable's registrar compiled its own records into "the first actual Mortality Experience of Assured Lives which had been available to Actuaries," and it came out more favourable than Northampton. Every earlier table described a general population; this one described the people who actually buy assurance. The society that spent a century refusing to price by age had, as a by-product, been accumulating the dataset the pricing needed.

Three people who anchor more than this cluster

• observation-halley-founded-parallax-and-actuarial-pricing-both-vindicated-posthumously — Halley authored two founding methods, transit-of-Venus parallax and annuity pricing, and both were adopted only after his 1742 death. The lag documented here is not an accident of insurance; it is the second instance of the same shape in the same man's work.
• observation-richard-price-anchors-actuarial-and-political-philosophy-lineages — Price is load-bearing in both actuarial mathematics and the founding argument of modern political philosophy.
• observation-galloway-is-a-third-astronomer-actuary-instance — added 2026-07-28: Thomas Galloway, the registrar who deepened Pensam's 1811 mortality analysis in 1841 (see the Pensam bullet above), turns out to have run a concurrent, Royal-Medal astronomical career — a third and tighter instance of the same one-person-two-lineages shape, this time simultaneous rather than sequential.

How the vault got this wrong, and how it found out

This cluster is unusual in carrying a full record of its own error, and that record is worth keeping wired together — it is the most transferable thing here.

• claim-tontine-coffee-house-dividend-proportional-claim-has-no-citation — the secondary sentence that put "in proportion to the shares those members owned" into the vault, carrying no inline citation. The likely route of the error: Walford uses "share" throughout to mean a share of the divided death-fund per claim — "death-share," "per share" — and it reads, to a modern eye, as an equity holding.
• claim-premature-negative-finding-amicable-dividend-was-an-unlocated-installment — the vault then declared the question unanswerable while the answer sat in the next installment of a source already in hand. A serialized article was treated as a single document, and every check run was a check on access when access was never the binding constraint.
• claim-walford-1885-essay-freely-hosted-installment-predates-amicable-society-1706 — the access finding that was true but became a stopping condition: Walford is free from the Institute and Faculty of Actuaries, and the first installment genuinely does end before 1706. What was missing was the instruction to increment the page range.

The general lesson is recorded at reflection-recurring-tool-summary-is-not-the-source, and the negative-finding half belongs to moc-argument-from-silence — specifically as an instance of the failure mode that MOC warns about, a silence trusted before the corpus was exhausted.

Entity hubs

The spine — entity-edmond-halley · entity-james-dodson · entity-richard-price · entity-thomas-galloway

The institution and its historians — entity-amicable-society-for-a-perpetual-assurance-office · entity-cornelius-walford · entity-institute-and-faculty-of-actuaries · entity-geoffrey-clark · entity-m-e-ogborn

Two of the three spine hubs reach outside this cluster, which is worth knowing before treating any of them as local: Halley also heads the vault's transit-of-Venus parallax material and is a live citation in a 2007 exoplanet paper, and Price also anchors the Burke / Revolution Controversy lineage. Dodson is the one who belongs wholly to this story. Clark and Ogborn are hubs for books the vault has never read; both are genuinely access-blocked, and both are now unnecessary for the questions they were held for.

Open threads

• DISCHARGED 2026-07-25 — the four unpromoted Walford findings this section previously listed are now claim-notes, wired into the sections above: the 1807 adoption, Pensam's 1811 experience, Hartley's 1713 embezzlement, and the 1711 policy sale.
• The 1711 priority claim is deliberately soft. claim-1711-amicable-policy-sale-possibly-first-life-policy-sale-on-record is held at seedling with an [unverified-priority] watch_flag: "probably the first instance" is Walford's own hedge, offered without a survey. Clark's Betting on Lives or a modern history of life-settlement markets would be the check.
• ANSWERED 2026-07-26 — the cluster's last open question, and the spine is weaker than it looked. The causal claim that Dodson's refusal produced his design is reported tradition, not documented fact: the ODNB hedges it ("It was said"), cites no source for it, and names Ogborn 1962 as its own authority; the freely-hosted JIA review of Ogborn explains Dodson's motivation without mentioning the Amicable. See claim-dodson-rejection-causation-is-reported-tradition-not-documented. The MOC's opening sequence still holds — refused at the gate, then built the alternative — but the therefore joining those two clauses is a story the sources repeat rather than one they evidence, and the sections above are written to state the sequence and not the causation.
• Genuinely still unread: Ogborn 1962 (HathiTrust catalog 001741155, behind a Cloudflare challenge no institutional IP clears) and Clark 1999. No open question now depends on either. Dodson's own First Lecture on Insurances is unpublished per the ODNB, so the primary route would require manuscript work.
• DISCHARGED 2026-07-26 — the three missing hubs are built: entity-edmond-halley, entity-james-dodson, entity-richard-price. Two of them turned out to span clusters rather than sit inside this one, which is the argument for having built them: Halley also heads the transit-of-Venus parallax cluster, and Price also anchors the Burke/Revolution-Controversy material.