Appendix J: A Timeline of Insurance and Underwriting

Nothing on your desk is new. The wind deductible you set this morning, the loss run you read before the application, the slip the broker walked into your inbox, the model that scored the account and the override you wrote over it — every one of them is the residue of three thousand years of merchants, mathematicians, and burned cities trying to solve the same problem: how to make a catastrophe for one survivable for all. Chapter 2 tells this story as a narrative, with the why of each turn; this appendix is the timeline you can scan when you need the when. Read them together. The chapter explains how a coffeehouse became a market and how a mortality table turned a wager into a discipline; this list puts those turns in order, adds the modern catastrophes that reshaped the field after the chapter's main arc, and points forward to where Chapter 36 picks the story up — the underwriter of 2035, climate repricing, and the algorithm moving from advisor to author.

A word on honesty before the dates, in the spirit of the citation rules that govern this whole book. Every statute, institution, regulator, and public event named here is real and described qualitatively. We give the year of an event we are confident of; where the precise date of an ancient practice or a gradual development is genuinely uncertain, we name the era rather than invent a false precision. We name no casualty totals, no insured-loss dollar figures, and no market sizes — those vary by source, and a timeline that invents one to look authoritative has failed at the one thing a reference is for. For how much, the chapters send you to the actual source.

📋 At the Desk Read a timeline the way you read a loss run: for the pattern, not the line items. Three patterns recur, and they are the ones Chapter 2 asks you to carry. First, catastrophe drives reform — the Great Fire, the 1906 earthquake, Andrew, September 11, Katrina each forced a change that prosperity never would. Second, the math arrives late and changes everything — pooling worked for centuries before anyone could price it, and every leap in pricing power (mortality tables, credibility, catastrophe models, predictive models) reset what was insurable. Third, scandal writes regulation — the tontine collapse, the solvency failures, the antitrust fight each produced the rules you now work inside. Watch for all three below.

The deep history of insurance is the history of two ideas — sharing a loss and transferring a risk for a price — long before anyone separated insurance from lending, or pricing from guesswork (§2.1).

c. 3rd–2nd millennium BCE (era) — Bottomry and the Babylonian codes. The oldest ancestor of insurance is the bottomry loan: a maritime advance secured by a ship (its "bottom") that the borrower need not repay if the vessel is lost, the extra charge above ordinary interest functioning as a premium for the risk. The practice is genuinely ancient; legal codes of the Babylonian and broader Mesopotamian world — the tradition associated with Hammurabi's code among them — already addressed loss, liability, and the cancellation of a debt when a venture failed. The exact origin is lost to the era, not pinned to a year, and we leave it there. (See §2.1.)

c. 1st millennium BCE (era) — General average and the Rhodian sea law. Mediterranean maritime custom developed general average: the principle that a loss deliberately incurred to save the whole venture — cargo jettisoned to lighten a foundering ship — is shared proportionally among everyone whose property the sacrifice saved. The rule is traditionally traced to the sea law of the island of Rhodes and was later absorbed into Roman law. It is risk pooling expressed as a legal and moral rule rather than a contract, and it is still invoked in modern ocean-marine insurance today. (See §2.1; the term recurs in Chapter 26.)

Antiquity through the Middle Ages (era) — Burial societies, guilds, and mutual aid. Across the classical world and medieval Europe, people pooled small contributions against shared misfortunes — Roman burial societies that covered funeral costs, and later the craft and merchant guilds that supported members through fire, shipwreck, theft, and death. These were mutual aid, not commercial insurance, but they are the direct ancestors of the mutual insurer (Chapter 3): owned by their members, funded by the many, paying the unlucky few.

Part 2 — The birth of the modern industry (1600s–1700s)

Two events in a single century created what we would recognize as insurance: a city burned, and a coffeehouse became a market. The mathematics that would let either one be priced arrived in the same era (§2.2–§2.4).

1666 — The Great Fire of London. The fire began in a bakery on Pudding Lane in the early hours of September 2, 1666, and over several days destroyed a vast swath of the medieval City of London. It is the hinge of the whole story. The peril of fire was nothing new; what was new was the response. The disaster proved that a dense stock of buildings could burn together — a correlated catastrophe — and the rebuilding of London afterward created, for the first time, a large stock of broadly similar brick-and-stone structures that could be classified, inspected, and pooled. (See §2.2.)

1680s (era) — The first fire-insurance companies and inspection-based selection. In the years after the Great Fire, the first true fire-insurance offices were founded in London — the enterprise associated with Nicholas Barbon is conventionally cited as among the earliest. These offices invented two things the underwriter still lives by: selection (they would not insure every building at the same price) and prevention as the insurer's business (some maintained their own fire brigades and marked insured buildings with fire plaques). Fire insurance is the birthplace of inspection and of pricing graded to the physical risk — the ancestor of COPE (Chapter 9). (See §2.2.)

Late 17th century (era) — Edward Lloyd's coffeehouse. Edward Lloyd kept a coffeehouse in late-17th-century London (near the Thames, later on Lombard Street) that became the gathering place for those with news of ships — which had sailed, which had arrived, which were overdue — and therefore for those who wanted to insure voyages and those with capital to back them. A merchant or shipowner brought the risk; individuals wrote their names under the description of the risk, accepting a share of it — which is the literal origin of the word underwriter (one who writes under a risk). The coffeehouse grew into Lloyd's of London: not an insurance company but a marketplace where syndicates, backed by members' capital, subscribe to risks. The broker-underwriter structure you work inside every day was defined here. (See §2.3; Lloyd's recurs in Chapters 3 and 27.)

1693 — Halley's mortality table. The astronomer Edmond Halley — of the comet — constructed one of the first genuine life tables, using carefully recorded birth and death registers (from the city of Breslau) to estimate the probability that a person of a given age would die within the year. That probability is exactly the frequency a life underwriter needs. Halley's table is the moment mortality became something you could calculate rather than guess — the seed of actuarial science. (See §2.4.)

1762 — Equitable Life and the actuarial revolution. The Society for Equitable Assurances on Lives and Survivorships ("Equitable Life"), founded in London in 1762, is generally regarded as the first life-insurance company to price policies on a sound mathematical basis — to use mortality data, charge a level premium computed by men such as James Dodson, sort applicants into classes by the factors that predict mortality, and hold reserves adequate to its long-term promises. This is the rise of actuarial science: it converted life insurance from a wager into a discipline and built the quantitative spine of everything in pricing that follows (Chapters 10, 11, 17, 32). (See §2.4.)

Part 3 — The American story (1700s–1900s)

The American industry grew from the same two roots — mutual aid and the marine/fire market — and then learned, the hard way, that it needed both mathematics and regulation (§2.5–§2.6).

1752 — The Philadelphia Contributionship. America's oldest continuously operating property insurer, the Philadelphia Contributionship for the Insurance of Houses from Loss by Fire, was founded in 1752 with Benjamin Franklin among its organizers. A mutual (the policyholders owned it), it priced for hazard from the start — famously declining or surcharging riskier buildings such as those crowded by trees — and so carried the inspection-and-selection logic of London's fire offices into the New World. (See §2.5.)

1792 — The first American stock insurer. The Insurance Company of North America (INA) was founded in Philadelphia in 1792, an early American stock insurer (owned by shareholders, supplying capital for profit) writing marine and then fire risks — a marker of the industry's shift from mutual aid toward capitalized, for-profit carriers (the stock insurer of Chapter 3). (See §2.5.)

19th century (era) — Industrialization and the multiplication of lines. The nineteenth century turned a marine-and-fire trade into an industry. Steam power, railroads, factories, and cities created new perils and new lines — boiler and machinery insurance (the ancestor of equipment breakdown, Chapter 19) emerged from catastrophic boiler explosions, accompanied by inspection as a condition of coverage; accident and casualty lines grew with industrial injury; and the spread of workers' compensation statutes in the early twentieth century created the statutory, no-fault system of Chapter 22. The pattern repeats: a new technology creates a new exposure, and underwriting invents the line to carry it.

1861–1865 (era) — The Civil War and the federal income tax precedent. The period is a minor but real marker: wartime mortality stress-tested young life insurers, and the era's federal revenue measures foreshadowed the tax questions that would later collide with insurance regulation.

1871 — The Great Chicago Fire. A conflagration destroyed much of Chicago in October 1871, bankrupting many fire insurers that had concentrated their exposure in one burnable city. It is an early, brutal lesson in accumulation risk (Chapter 29) and the reason solvency and concentration discipline became survival skills, not refinements. (See §2.6.)

1905–1906 — The tontine collapse and the Armstrong investigation. Through the late nineteenth century, American life insurers sold enormous volumes of tontine (deferred-dividend) policies — named for the seventeenth-century financier Lorenzo Tonti — in which lapsing or dying policyholders forfeited their share to the survivors. The product was wildly popular and dangerously opaque, and the funds it accumulated bred misconduct. The Armstrong investigation in New York (1905) exposed serious abuses in the management of these funds and led to sweeping reforms — restrictions on deferred-dividend designs and a new architecture of policyholder protection — that shaped American life insurance and its state regulation permanently. The tontine is the book's standing parable: a practice that is profitable and popular is not therefore sound. (See §2.4 and §2.6.)

1906 — The San Francisco earthquake and fire. The April 18, 1906 earthquake and the fire that followed devastated San Francisco and produced one of the largest insured losses of its era. It hardened the line between earthquake shake and fire following in policy wording (a distinction that still matters in Chapter 15), bankrupted under-reserved insurers, and burnished the reputations of those — including major overseas carriers — that paid in full. Catastrophe, again, rewriting both the forms and the market. (See §2.6.)

1911 — The Triangle Shirtwaist Factory fire. The March 25, 1911 factory fire in New York City, which killed scores of garment workers behind locked doors, was a turning point for workplace-safety law and helped accelerate the spread of workers' compensation statutes across the states — a reminder that the social function of insurance and the law that shapes it move together (Chapter 22).

Part 4 — Regulation, the federal question, and the postwar order (1940s–1990s)

The mid-twentieth century settled the constitutional question of who regulates insurance and built the modern regulatory machinery — then watched a new generation of catastrophes test it (§2.6).

1944 — United States v. South-Eastern Underwriters Association. For decades it had been assumed that insurance was not "commerce" subject to federal regulation. In 1944 the U.S. Supreme Court held in the South-Eastern Underwriters case that the business of insurance was interstate commerce and therefore reachable by federal law — including the antitrust laws — throwing the long-settled system of state regulation into doubt overnight. (See §2.6 and Chapter 4.)

1945 — The McCarran-Ferguson Act. Congress responded the next year with the McCarran-Ferguson Act (1945), which returned the regulation and taxation of the business of insurance to the states and exempted it from most federal law — including, with limits, the antitrust laws — to the extent the business is regulated by state law. McCarran-Ferguson is the reason insurance in the United States is regulated state-by-state to this day, the foundation under everything in Chapter 4, and a Tier-1 statute this book stands behind by name. (See §2.6; defined in full in Chapter 4.)

Postwar (era) — The NAIC and the model-law system mature. The National Association of Insurance Commissioners (NAIC) — the coordinating body of the state regulators — long predates this period, but the postwar order is when the model law mechanism became the engine of national consistency without national control: the NAIC drafts a model, the states adopt versions of it, and uniformity emerges from the bottom up. This is the architecture behind risk-based capital, market conduct, and most of the rules you file under (Chapters 4 and 28).

1968 — The National Flood Insurance Program (NFIP). After private insurers had largely withdrawn from flood — a peril too correlated and too adversely selected to write profitably without intervention — Congress created the NFIP in 1968, a federally backed program (later administered by FEMA) that writes flood coverage, maps the nation's flood zones, and ties coverage to community floodplain management. It is the canonical example of a public-private response to an insurability gap, and of its limits: capped, historically subsidized, and under-purchased, it leaves most flood loss uninsured — the protection gap at its starkest (Chapters 15 and 30).

1970s–1980s (era) — The liability and asbestos crises. A wave of long-tail liability losses — above all asbestos and environmental (pollution) claims arising from exposures decades earlier — produced losses the original underwriters never imagined and never priced, and helped drive a severe liability insurance crisis in the mid-1980s. It is the permanent lesson of the long tail (Chapter 21) and of the occurrence trigger's reach across time: you can be paying tomorrow for a policy you wrote a generation ago.

1980s (era) — Risk-based capital and modern solvency monitoring take shape. A run of insurer insolvencies pushed the NAIC toward formal risk-based capital (RBC) standards — a formula that ties required capital to the specific risks on a carrier's balance sheet, with action levels that trigger regulatory intervention (Chapter 28). Solvency stopped being a matter of judgment and became a matter of measured capital adequacy.

Part 5 — The catastrophes that reshaped the modern field (1992–2005)

Four events in thirteen years remade property-catastrophe and specialty underwriting. Each revealed that the existing tools had badly underestimated what a single event could do — and each forced a permanent change in how the risk is measured, priced, and financed.

1992 — Hurricane Andrew. Andrew struck South Florida in August 1992 as a compact, intense hurricane and produced an insured loss far beyond anything the industry's models had contemplated. It bankrupted multiple insurers, drove others from the Florida market, and exposed the prior generation of pricing tools as dangerously inadequate for correlated catastrophe. Andrew is the event that made modern catastrophe modeling (Chapter 30) — simulating tens of thousands of plausible events against a specific portfolio — a non-negotiable part of property underwriting rather than an academic exercise. It also accelerated the rise of Bermuda as a catastrophe-reinsurance center. (See Chapter 30; referenced throughout as a Tier-1 event.)

1994 — The Northridge earthquake. The January 1994 earthquake in the Northridge area of Los Angeles produced an insured loss that stunned the residential-earthquake market and led insurers to retreat from writing earthquake coverage bundled with homeowners policies in California — a withdrawal that prompted the state to create a public earthquake authority to keep coverage available. Northridge is the seismic companion to Andrew: a single event resetting an entire line's appetite and the public-private structure behind it (Chapter 15).

2001 — September 11. The terrorist attacks of September 11, 2001 produced an unprecedented, cross-line catastrophe — property, business interruption, aviation, workers' compensation, life, and liability losses all from a single correlated event — a scenario no one had modeled or priced as a single accumulation. It hardened the entire commercial market, forced terrorism out of standard coverage as an explicit exclusion, and led to a federal backstop for terrorism risk (the program established the following year and renewed since), because the private market judged the peril uninsurable on its own. September 11 is the standing case for the limits of diversification and for catastrophe that ignores the boundaries between lines (Chapters 27, 30, and 36). (Tier-1 event.)

2005 — Hurricane Katrina. Katrina struck the Gulf Coast in August 2005, and the failure of the New Orleans levees turned a severe hurricane into one of the costliest insured catastrophes in U.S. history. Its defining underwriting legacy is the wind-versus-water dispute: standard homeowners policies cover wind but exclude flood, and Katrina's surge produced years of litigation over which peril caused which loss — a coverage fault line the industry now addresses head-on in wording and in the NFIP relationship (Chapter 15). Katrina also stress-tested catastrophe models on a storm-surge dimension they had underweighted, and pushed reinsurance pricing and capital sharply. (Tier-1 event.)

Part 6 — The modern era: data, regulation, and the climate turn (2008–present)

The most recent two decades layered a financial crisis, a health-care revolution, and a technology transformation onto a property market now being repriced by the physical climate (Chapters 18, 31–36).

2008 — The financial crisis and AIG. The crisis is an insurance story in one crucial respect: AIG, a giant insurer, was brought to the brink not by its core underwriting but by a financial-products unit writing credit derivatives — a reminder that an insurer's risk can come from somewhere other than its policies, and the backdrop for enterprise risk management (ERM) and the ORSA of Chapter 28. The episode also sharpened the debate over whether insurance regulation should remain purely state-based (it largely did, under McCarran-Ferguson) or acquire a federal dimension for systemically important firms.

2010 / 2014 — The Affordable Care Act. The Affordable Care Act (ACA) was enacted in 2010, and its central market reforms took effect in 2014: guaranteed issue and adjusted community rating abolished most individual medical underwriting, replacing the underwriter's selection-and-pricing role with risk adjustment, a medical-loss-ratio rule, and enrollment controls. It is the clearest case in the book of a public policy choice removing underwriting from a market — and of the adverse-selection mathematics that then had to be managed by other means (Chapter 18). (Tier-1 statute.)

2010s (era) — Catastrophe modeling matures and telematics arrives. The decade saw catastrophe models become richer (multi-peril, climate-conditioned) and, in personal auto, the arrival of telematics and usage-based insurance at scale — pricing from measured behavior rather than static proxies (Chapter 14). The raw material of underwriting began shifting from what the applicant tells you to what independent data shows.

2010s–2020s (era) — The InsurTech wave. A surge of technology-driven companies — digital MGAs, embedded insurance, parametric products, API-driven distribution, and full-stack digital carriers — set out to rebuild distribution, products, and underwriting (Chapter 34). The wave delivered real gains in speed and reach and a set of hard public lessons: several high-profile digital carriers discovered that writing business fast is not the same as writing it profitably, and that the combined ratio judges an algorithm exactly as it judges a human. (Specific company results are kept qualitative; see Chapter 34.)

2017–present (era) — The California wildfire crisis and the climate turn. A series of severe wildfire seasons beginning around 2017 produced insured losses that broke the prior models' assumptions and led major insurers to restrict or stop writing homeowners coverage in parts of California and other wildfire-exposed regions — an availability-and-affordability crisis that pushed more homeowners into the state's insurer of last resort. This is the leading contemporary example of climate repricing: not a one-time adjustment but a moving baseline, in which yesterday's hundred-year event becomes more frequent and the law of large numbers cannot settle on a stable expected loss. It is the hard physical truth at the center of Chapter 36 — the future you cannot invent away. (See Chapter 36; the wildfire insurability crisis is a Tier-1 reference point.)

2020 — COVID-19 and the business-interruption litigation. The pandemic produced a wave of business-interruption claims and litigation testing whether a virus and a government shutdown order triggered property-based BI coverage that generally requires physical loss or damage. The episode is a live lesson in policy-language precision (Chapter 5) and in how a genuinely novel, correlated, non-physical peril stresses contracts written for a different world. (Tier-1 event; outcomes vary by jurisdiction and wording and are kept qualitative.)

2020s (era) — Generative AI and the algorithm as author. The most recent turn is the arrival of powerful predictive and generative AI in the underwriting workflow — models that not only score a static submission (Chapter 32) but read, summarize, and draft the underwriter's work, and increasingly monitor a risk continuously rather than once at the point of sale (Chapters 31 and 36). The book's whole argument lives in this entry: the algorithm is moving from advisor to author, and the enduring skill is the one this textbook exists to teach — knowing what the model can and cannot see, and defending the judgment that overrides it. (See Chapter 36; "AI as co-pilot, not autopilot.")

How to use this timeline

🔍 Check Your Understanding 1. Name three catastrophes from this timeline and the specific change each one forced in how risk is measured, priced, or financed. (Andrew → catastrophe modeling; September 11 → cross-line accumulation and a terrorism backstop; Katrina → wind-versus-water wording and surge modeling are three of several.) 2. What two developments most reduced uncertainty about price — and roughly when did each arrive? (The mortality table and Equitable Life in the 17th–18th centuries; modern catastrophe and predictive models in the 1990s–2020s.) 3. Which entry best illustrates a public policy choice removing underwriting from a market, and which best illustrates the limits of insurability? (The ACA removes individual medical underwriting; the California wildfire crisis and the NFIP mark the limits of insurability and the protection gap.)

This appendix is deliberately a skeleton — the years and the turns, not the reasoning. The reasoning is in the chapters, and the cross-references above are the doorways back to it. When you want the story of why a coffeehouse became the model for the market you work in, or why a burned city had to invent the inspection you order on every property risk, read Chapter 2 end to end. When you want to know where the line goes next — continuous underwriting, climate repricing, and the underwriter who can bridge judgment and analytics — read Chapter 36. The pattern that connects them is the one worth memorizing: catastrophe forces reform, the math arrives late and resets what is insurable, and scandal writes the rules. You are standing on all three, and adding the next entry to this list yourself.

⚠️ Underwriting Trap Do not read a timeline as progress — as if each era simply improved on the last and the modern underwriter has nothing to learn from the bottomry lender or the coffeehouse syndicate. The opposite is true. Every tool on this list solved a real problem and created a new way to be wrong: the mortality table made it possible to be precisely wrong about a class, the catastrophe model made it possible to trust a simulated number past what it can support, and predictive AI makes it possible to launder an old bias through new math (Chapter 35). The history does not end with the algorithm. It ends — for now — with you, holding the judgment the machine cannot, which is exactly where it began.