What it means
How bad could it get? Underwriting needs a number, not a shrug, and probable maximum loss is the industry's disciplined answer for a single event on a single risk.
The PML estimates the worst loss under expected conditions: the fire with sprinklers working, the hurricane with the roof strapped, not the Hollywood scenario where everything fails at once. The concept is sometimes framed against its gloomier cousin, the maximum foreseeable loss, which assumes protection systems fail, while PML assumes they perform.
Catastrophe modelling formalised the idea, and the NAIC's overview of catastrophe models describes exceedance probability curves from which probable maximum losses at chosen return periods can be read directly. A 250-year PML, for instance, is the loss level with a 0.4% annual chance of being exceeded.
Insurers pick their reference return period based on their risk appetite and regulatory capital rules. The number drives decisions up and down the chain: property insurers price with it, reinsurers structure treaties around it, lenders demand coverage sized to it, and regulators test solvency against it.
Insurers do not carry their gross PML alone, since reinsurance treaties are layered precisely around these numbers, with each layer priced against the probability of being reached. The concept also disciplines buyers, because a firm that knows its factory's PML can weigh sprinklers against premiums on the same scale instead of treating safety spending and insurance as separate budgets.
Models being models, PMLs embed judgement about construction, geography, and correlations, and the 2011 Tohoku earthquake taught the industry that tails can be fatter than the curves assumed. Climate change complicates the curves further, because when the past stops describing the future, return periods drift and yesterday's 250-year number may be tomorrow's 100-year one.
Regulators lean on the same arithmetic, with stress tests and capital standards asking, in effect, whether the company survives its PML with room to spare. For a non-finance reader, PML is the difference between insurance as a comfort word and insurance as arithmetic.
Somebody computed the worst plausible Tuesday and priced the promise to survive it.
In practice
Real-world examples.
Example
A lender requires earthquake coverage sized to the building's PML, not just its replacement cost, before approving the mortgage. The borrower commissions an engineering study to support the figure.
Example
An insurer reads its 200-year PML from the catastrophe model's exceedance curve and buys reinsurance above that point. The curve, not a hunch, drew the line, and the board records the return period it chose.
Example
After a model update raises flood estimates, a carrier's coastal PML jumps 20% and its pricing follows. Underwriters restrict new coastal policies until the extra reinsurance cost is built into premiums.
Formula
Calculation
From catastrophe models: PML at return period T is the loss level whose annual exceedance probability is 1/T. A 100-year PML is the loss with a 1% chance of being topped in any year, read off the exceedance probability curve.
Worked example. A 250-year PML has an annual exceedance probability of 1 / 250 = 0.4%. A 100-year PML has 1 / 100 = 1%. Over a 30-year holding period, the chance of at least one 100-year event is 1 - (0.99 to the power of 30) = 1 - 0.74 = about 26%, so a "rare" loss is far from remote over the life of a mortgage. If the model puts the 100-year PML for a fictional warehouse at $12,000,000 and the owner holds $10,000,000 of cover, the owner is exposed to a $2,000,000 gap in that scenario.Case study
Seen in the real world.
This case study is fictional and illustrative. A made-up property insurer in Florida runs its catastrophe model on a coastal homeowners' book. The output shows a 100-year PML of 420 million dollars and a 250-year PML of 610 million. Its surplus is 500 million, so a 250-year storm would break the company outright.
Management acts on the arithmetic. It buys reinsurance covering losses above 300 million up to 650 million, lifting its survivable event beyond the 250-year level, at a premium of 38 million a year. The board debates the cost until the CFO puts it plainly: the reinsurance premium is the price of the company's own 250-year promise to policyholders. When a major hurricane lands two years later with a modelled 380-million-dollar gross hit, the reinsurer takes 80 million above the retention, the insurer's surplus holds, and the PML chart that drove the purchase is framed in the boardroom.
Watch out
Common mistakes.
- Confusing PML with maximum possible loss; PML assumes protection systems work, while worst-conceivable scenarios assume everything fails at once. The two numbers answer different questions.
- Treating the model output as fact; PMLs embed assumptions about construction and climate that real events periodically humble.
- Sizing coverage to average losses; the point of PML is the tail, and solvency depends on surviving the bad year, not the typical one.
Questions
People also ask.
What is probable maximum loss?
The largest loss reasonably expected from a single event assuming mitigation works, used to size coverage, reinsurance, and capital.
How is it calculated?
Catastrophe models generate exceedance probability curves over thousands of simulated events, and the PML is read at a chosen return period, such as 100 or 250 years.
What is the difference from maximum foreseeable loss?
PML assumes sprinklers, straps, and defences perform; the maximum foreseeable loss assumes they fail, producing the grimmer number.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related