Fire, Coffee, and Mathematics, in brief
The 4-minute version of Fire, Coffee, and Mathematics.
Joel R. Singh
Underwritten
2026-09-10
In the small hours of 2 September 1666, a fire caught in the king's bakery on Pudding Lane. London was built of timber and pitch, its upper floors leaning over the lanes, and the summer had been dry. The fire burned for four days, destroyed roughly thirteen thousand houses, and left tens of thousands with nothing.
Samuel Pepys buried his wine and his parmesan cheese in the garden. That one detail tells you how completely ordinary life came apart.
Fire changes everything
Out of those ashes came fire insurance. Nicholas Barbon, a physician turned speculator, sold fire cover after the blaze and formalized it as the Fire Office by 1680.
What matters is what he did beyond collecting premiums. Barbon ran his own fire brigade and gave every insured household a metal badge to nail up front. When a fire broke out, the brigade looked for the badge and knew whether this was a house they were paid to save.
An insurer who has to write the check has every reason to stop the fire first. Barbon acted on that with men and buckets, and rivals followed. The idea never left us: it is why a sprinkler lowers your premium today.
The coffeehouse where "underwriter" was born
The first record of Edward Lloyd's coffeehouse dates to February 1688. Lloyd never insured a voyage. He ran a room, and it was the room where the maritime trade gathered.
A broker wrote a voyage's details on a slip of paper: the ship, her master, the route, the cargo, the season. He carried it around to the men with money. A backer who liked the risk read the terms, decided what share he would carry, and wrote his name underneath.
That signature under the line is where the word underwriter comes from. An underwriter is the person who examines a risk and decides on what terms to accept it.
The clever part came when several men signed one slip. A dozen names might sit beneath one voyage, each taking a fraction. Nobody was exposed to all of it. That gathering grew into Lloyd's of London, still a marketplace rather than a company.
Death becomes arithmetic
Everything so far ran on judgment. Nobody could say what a fair premium actually was.
That changed in 1693. The astronomer Edmond Halley, the one with the comet, published a study of the dead. A pastor named Caspar Neumann had recorded every birth and death in the city of Breslau, with ages. Halley turned those registers into one of the first mortality tables. It showed how many of a group born in one year could be expected to be alive at each later age.
Any single death is unknowable. Across ten thousand lives, death becomes a smooth curve, and a curve is something you can price. Halley used his table to calculate a fair price for a life annuity. The oldest question in the trade finally had an answer you could compute.
The level premium, and the first actuary
Charge a person each year exactly what that year's risk of death costs, and the premium starts cheap while he is young and turns crushing when he is old and needs it most.
James Dodson's answer in the 1750s was the level premium: the same steady amount every year. The early years deliberately overcharge, and that surplus grows into a reserve that pays for the expensive later ones. It is the engine behind nearly every whole-life policy since.
Dodson had been turned away by an insurer that would not take anyone over forty-five. He died in 1757, but the men he gathered carried the plan forward. In 1762 the Equitable opened in London, the first life insurer priced on a real actuarial footing. Its leader gave the officer running the mathematics a title nobody had worn: the actuary.
Franklin's rule
In 1752 Benjamin Franklin and his neighbors founded the Philadelphia Contributionship. It is the oldest property insurer in the United States and still writes cover today.
Two things about it reach into the present. It was a mutual, owned by its policyholders, which runs straight back to the Roman burial societies of Part 1. And it treated inspection as a condition of cover. It looked at the houses it insured, priced by how they were built, and refused hazards it judged too dangerous at any price.
That habit is not a curiosity. It is the same principle at work when an insurer requires fire doors, or encrypted records, or documented testing of an AI system, before it will write the policy.
What comes next
By 1762 the machinery was assembled. There was pooling, the priced premium, loss prevention with money behind it, mutual ownership, and the mathematics to hold it together. Everything after is that machine meeting new dangers. Boilers, then automobiles, then data breaches. The newest risk does not break down or catch fire. It makes decisions, and sometimes it makes them wrong.
That is Part 3, Insuring the Machine Age.
Works Cited
- 1The Great Fire of London, Thomas Farriner, Pudding Lane, and the scale of roughly 13,000 houses lost
- 2Samuel Pepys burying his wine and parmesan cheese https://www.pepysdiary.com/diary/1666/09/04/
- 3Nicholas Barbon and the Fire Office
- 4Edward Lloyd's coffeehouse, first recorded mention February 1688 on Great Tower Street
- 5Edmond Halley's mortality table
- 6Lorenzo de Tonti, the 1653 tontine proposal to Cardinal Mazarin
- 7James Dodson, the level premium, the Amicable Society's age-45 refusal, and Dodson's death in 1757
- 8The Society for Equitable Assurances on Lives and Survivorships (the Equitable), founded 1762, and its use of Northampton mortality data
- 9Edward Rowe Mores coining "actuary" (1762)
- 10The Philadelphia Contributionship for the Insurance of Houses from Loss by Fire, founded 1752, oldest property insurer in the United States https://1752.com/
---
Informational only. This essay is a general history and is not insurance advice, legal advice, or a recommendation of any policy. Dates and attributions follow the conventional historical record and are approximate where the sources themselves are. Research and writing by Joel R. Singh for iSinghLabs Inc.