Insure the Tail, Not the Average

Every policy is a losing bet on average

An insurer collects premiums, pays claims, covers its expenses and keeps a margin. Add those up and the expected value of any policy you buy is negative — otherwise the insurer would not sell it. That is not a scandal; it is what pooling costs. So “is this insurance a good deal on average?” is the wrong question, and the household that asks it will end up under-insured on the tail and over-insured on the trivial. The right question is the shape of the loss. A $1,200 phone replacement is affordable and frequent: paying a premium for it means paying the insurer’s margin on a rounding error, over and over. A permanent loss of a working parent’s income is neither affordable nor recoverable, and no household can self-insure it out of savings — which is why disability and term life cover are the two most valuable policies a young family can own, and often the two they do not have. This gives a clean decision rule. **Retain** risks whose worst case you could pay from the buffer and absorb within a couple of years. **Transfer** risks whose worst case would change the course of the plan — and transfer them with the largest deductible you can comfortably carry, because the deductible is the part of the loss you have chosen to self-insure and it is priced accordingly. Which is which — Retain — small, frequent, affordable: phone cover, extended warranties, small deductibles · Transfer — large, rare, ruinous: term life, long-term disability, umbrella liability · The rule for the deductible: as high as the buffer can carry, because it is priced · What insurance is not: an investment, a savings plan, or a way to beat the odds ← A policy that bundles insurance with a savings or investment component — whole life with a cash value, a variable annuity inside a retirement account — usually charges for both jobs and does neither cheaply. Buy term insurance for the tail and keep the investing separate.

The tails a household actually has

Four risks do most of the damage, and each has a cheap instrument. **Loss of income from death**: term life insurance, sized as the years of income dependents need plus debts, minus what the household already has — and bought level for exactly those years, because that is when the cover is needed. **Loss of income from illness or injury**: long-term disability, the risk a household is far more likely to meet than death during its working years, and the one most often skipped. An own-occupation definition matters there, because a policy that only pays when you cannot do any work is worth much less than one that pays when you cannot do your work. **Liability**: an umbrella policy of a million dollars costs a few hundred a year and sits on top of the home and auto policies. It is the cheapest tail cover available, because the events are rare and the consequences are unbounded — one serious accident with a judgment against you can reach future income and savings, which is exactly the kind of loss a buffer cannot pay. **Health**: the number to read is the out-of-pocket maximum, because that is the size of the tail you are retaining. If the maximum is a number you could pay from the buffer, the plan is sound; if it is not, the plan is exposed. Two features are worth comparing across policies rather than the headline price: the **deductible** (what you retain, and where the premium saving lives) and the **definition and duration of cover** (how the policy behaves in the case it exists for). Cheap life cover that expires before the children leave school, or disability cover that pays for two years on a thirty-year mortgage, is a discount on something that does not do the job. Four tails, four instruments — Death while dependents rely on your income: level term life, sized to the gap and the years · Illness or injury stopping your work: long-term disability, own-occupation if possible · A liability judgment larger than your assets: a $1m umbrella on top of home and auto · A health catastrophe: cover whose out-of-pocket maximum you could pay ←

Where the money leaks

Insurance is sold with a margin, so the products with the smallest claims and the largest emotional pull carry the best margins for the seller. Extended warranties, phone protection, trip cancellation on a weekend away, mortgage payment protection: each is a small, frequent and affordable loss that a household can pay from its buffer, priced to be profitable for the insurer precisely because the loss is manageable. The pattern to look for is a product that insures the middle rather than the tail. Insurance makes sense when the worst case is ruinous; it makes little sense when the worst case is annoying. The household that declines the small covers and holds a slightly larger buffer comes out ahead on both counts — cheaper premiums, and a buffer that covers several small losses at once instead of one narrow event. There is a second, subtler leak: duplication. Life cover and mortgage protection insurance can cover the same risk; critical-illness and disability policies can overlap; an accident policy on top of a health plan with a low maximum adds very little. Before buying another policy, list what is already covered and at what level, because the cheapest tail cover is the one you already own and had forgotten. The same household, two philosophies — Buys small covers, low deductibles, bundled products: high premiums, narrow payouts, thin tail cover · Self-insures the small, high deductibles, term and umbrella: lower premiums, wider buffer, real tail cover ← The buffer and the deductible are the same decision seen from two directions: what you retain should be a number you could pay without changing the plan.

Insurability is a perishable asset

Insurance has a property that almost no other financial decision shares: the ability to buy it can disappear. A policy is priced and issued against your health at the moment you apply, and a diagnosis, an injury or a change in occupation can make cover unavailable or permanently more expensive. That makes insurability an asset in its own right, and one that decays without anything appearing on a statement. The mechanism is underwriting. An insurer assesses a risk at application and then, for most individual policies, cannot reprice it later for changes in health — which is precisely why the assessment is thorough at the outset. The consequence is asymmetric: a healthy applicant has a wide menu and pays the standard rate, a mildly impaired one has a narrower menu at a higher price, and an impaired one may have no individual market at all and depend on cover attached to an employer. The decision to buy cover when nothing has happened yet is therefore not pessimism, it is the only time the purchase is available on good terms. The practical consequence is that the ordering of the earlier reads — insure the catastrophic tails — has a time dimension. Disability cover and life cover should be put in place while the household is young and healthy, not at the point when the need becomes obvious, because by then the price and availability have moved. The same applies to long-term care cover, where the age at purchase is the dominant factor in the premium and where the policy terms themselves have shifted over time. There is a second, quieter version of the same point in employer-provided cover. Group policies are frequently cheaper than individual ones and are usually issued without individual medical underwriting, which is a real advantage. What they lack is portability: leaving the employer, or the employer changing its provider, ends or changes the cover, and someone who has relied on group disability or life cover for a decade may find that the individual version is unavailable at the age when they need it. The usual resolution is to hold the group cover as the base and to add individual cover while still insurable, rather than assuming the group policy travels with the person. The final piece is the point at which a household stops needing insurance, and it is worth stating because it prevents over-buying later. Term life cover exists to protect dependants from the loss of income during the years when they depend on it; as the mortgage is paid and the portfolio grows, the need falls, and a policy can be allowed to lapse rather than renewed. Disability cover is the mirror: its need persists as long as the household depends on earned income, which is a longer period than most people assume. The two tails move in opposite directions, and the review should treat them differently. • The ability to buy cover at a standard rate decays with health and is not recoverable. • Individual policies cannot generally be repriced later, which is why underwriting is thorough at issue. • Group cover is cheap and unportable; holding it alone is a hidden exposure. • The life-insurance need falls as the mortgage is paid; the disability need lasts as long as earned income does. A date worth putting in the plan: the next time the household’s health or employment situation will be reviewed — a job change, a birth, a mortgage. Those are the moments when cover needs re-examining, and they are also the moments when the insurability has not yet become an issue.

Sources

• Insurance as a transfer of risk rather than a betStandard risk-pooling arithmetic; FINRA investor education
• Life and disability cover: what a household actually needsConsumer Financial Protection Bureau; SEC investor bulletins on insurance products
• Umbrella liability coverInsurance Information Institute — personal liability guidance
• Out-of-pocket maximums in health plansHealthCare.gov — plan terminology and limits