1.1 Risk, Hazards, Perils, and the Law of Large Numbers
Key Takeaways
- Insurers cover only pure risk (loss or no loss); speculative risk (chance of gain) is uninsurable.
- A peril is the cause of loss; a hazard increases the chance or severity. Moral = dishonesty, morale = carelessness, physical = a tangible condition.
- Risk handling techniques: Sharing, Transfer, Avoidance, Reduction, Retention (STARR); insurance is transfer and a deductible is planned retention.
- The law of large numbers lets insurers predict aggregate losses across many similar units, making rates adequate, not excessive, and not unfairly discriminatory.
- Insurable risks should be Calculable, Affordable, Noncatastrophic, Homogeneous, Accidental, and Measurable (CANHAM).
Why Insurance Exists
Insurance is a financial mechanism that transfers the cost of a possible loss from one party (the insured) to another (the insurer) in exchange for a premium. It does not eliminate risk; it spreads the financial consequences of loss across a large pool of similar exposures so no single member is ruined. Roughly 30 percent of state exam questions in the national portion test whether you can correctly distinguish risk, peril, and hazard, because every later topic (perils-covered, exclusions, underwriting) builds on these three words.
Risk and Its Two Types
Risk is uncertainty about loss. Insurers care only about pure risk — situations with a chance of loss or no loss but never a chance of gain (a house burns or it does not). Speculative risk carries a chance of loss, no loss, or gain (gambling, stock trading, starting a business) and is not insurable. Memorize the distinction: if there is any upside, it is speculative and uninsurable.
Peril vs. Hazard
A peril is the cause of a loss — fire, windstorm, theft, collision, lightning. A hazard is a condition that increases the likelihood or severity of a peril. The exam tests three hazard categories:
| Hazard Type | Definition | Example |
|---|---|---|
| Physical | A tangible condition | Oily rags in a basement; icy steps; frayed wiring |
| Moral | A dishonest tendency to cause a loss | Arson for profit; faking a theft |
| Morale | Indifference/carelessness because insurance exists | Leaving keys in the car; not locking doors |
Trap: candidates confuse moral (intentional dishonesty) with morale (careless attitude). Morale = MORE relaxed attitude. Moral = a deliberate dishonest act.
A homeowner stops locking the front door because "the insurance will pay if anything is stolen." This attitude is an example of:
Handling Risk — STARR
There are five techniques to manage risk; remember STARR: Sharing, Transfer, Avoidance, Reduction, Retention. Insurance is the primary form of transfer. Avoidance means not undertaking the activity at all (never owning a pool). Reduction lowers severity or frequency (sprinklers, deadbolts). Retention means accepting the risk yourself — a deductible is planned retention. Sharing distributes risk among a group (a partnership, a reciprocal exchange).
The exam often pairs these with deductibles and coinsurance. A $1,000 deductible on a homeowners policy is retention of the first $1,000 of every loss; the insurer is only handling the layer above it.
The Law of Large Numbers
Insurers price coverage using the law of large numbers: as the number of similar, independent exposure units increases, the actual loss experience approaches the predicted (probable) loss experience. With ten houses, fire losses are wildly unpredictable; with one million houses, the annual loss rate is remarkably stable. This statistical reliability is what lets an actuary set a premium that is adequate (covers losses + expenses + profit), not excessive, and not unfairly discriminatory.
Worked example: If historical data show 4 fires per 1,000 insured homes per year, and the average fire loss is $50,000, the pure loss cost per home = (4 / 1,000) × $50,000 = $200. The insurer adds expense loading (say 35 percent) for a gross rate near $200 / (1 − 0.35) = $308 per home.
Elements of an Insurable Risk (CANHAM)
Not every pure risk is commercially insurable. A risk should be: Calculable (loss probability can be estimated), Affordable (premium is economically feasible), Noncatastrophic (not so large it bankrupts the insurer — pure war and nuclear losses are excluded for this reason), Homogeneous (large number of similar units), Accidental (fortuitous, outside the insured's control), and Measurable (definite in time, place, cause, and amount).
Trap: a single, unique exposure (one space shuttle, a celebrity's voice) violates the "large number of homogeneous units" element — those risks go to specialty/excess markets like Lloyd's, not standard insurers.
Notice how the elements interlock with everything above: "accidental" restates the pure-risk rule (no intentional or speculative losses); "noncatastrophic" is why war, nuclear, and flood are excluded from standard forms; "homogeneous" and "calculable" are what make the law of large numbers work; and "measurable" is why a loss must be definite in time, place, and amount before a claim is payable. Master these six and most of Chapter 1 becomes a single connected idea rather than isolated definitions.
Frequency vs. Severity
Underwriters separate two dimensions of loss. Frequency is how often losses occur; severity is how large each loss is. A fender-bender is high-frequency, low-severity; a total building fire is low-frequency, high-severity. The expected loss cost an insurer must fund is roughly frequency × severity. Risk-reduction measures target one or both: deadbolts and alarms cut theft frequency, while sprinklers and fire-resistive construction cut fire severity.
Catastrophe perils (hurricane, earthquake) are the dangerous combination of low frequency but extreme severity, which is why insurers cap aggregate exposure in a region and buy reinsurance.
Adverse Selection and Underwriting
Left unmanaged, the people most likely to suffer a loss are the most eager to buy insurance — this is adverse selection. Insurers counter it through underwriting: classifying applicants into homogeneous groups so each pays a rate matching its true expected loss. Tools include applications, inspections, loss-history reports (CLUE), credit-based insurance scores where permitted, and motor-vehicle records. Proper classification keeps the pool balanced and the law of large numbers reliable.
If high-risk insureds paid the same as low-risk insureds, the low-risk members would leave, losses would rise, and the program would spiral — the reason rates must be not unfairly discriminatory rather than identical for everyone.
An actuary observes that with 2,000,000 insured autos the annual collision loss rate stays within a narrow, predictable band each year, even though any single driver's loss is uncertain. This is a direct application of: