Life and disability insurance explained

Most working people own one asset larger than anything on their bank statement, and it appears on no balance sheet: the earnings they have not made yet. Someone at thirty with three decades or more of work ahead is carrying the present value of all of it, and by default that asset is uninsured.

Two events destroy it. Death ends the income and leaves whoever depended on it to carry on without it. Disability ends the income too, and leaves the person who lost it still needing to be supported. Life cover answers the first case, disability cover the second, and everything below follows from that split.

The asymmetry between them is the reason disability cover is not simply life cover for the living. A death removes an earner and a spender at once: the household loses income and also loses a set of costs. A long disability removes the earner and keeps the spender. Spending can rise rather than fall, and saving for retirement usually stops in the same week. Read against a household balance sheet, the disability case is often the worse of the two.

The test for buying either is the test that runs through the whole subject. Risk pooling turns a rare loss nobody could carry alone into a small known cost, and it is worth paying for precisely where the loss could not be absorbed. So ask what the loss would do to the people holding it. If it would be uncomfortable, savings can meet it. If it would change how they live, that is what a policy is for. Net worth is the usual measure of what can be absorbed, and for most people under forty it is small next to the earnings still to come.

Life insurance does not insure a person. It insures the money that stops when that person does, which makes the question factual: if you died tonight, would anyone face a shortfall they could not cover?

• Anyone living on your income. Children, a partner whose standard of living rests on what you earn, a parent you support.
• A partner who does unpaid work. A parent at home produces childcare and household labour that would have to be bought. No salary stops and a bill starts.
• Debt that somebody else signed. Joint mortgages, co-signed loans, a personal guarantee behind business borrowing.
• An estate that cannot be sold quickly. A farm, a private company, property, where a settlement cost falls due before the assets can be turned into cash.
• A business that would lose someone it cannot replace fast, which is what key person cover and a funded buy-sell agreement are for.

The other side of the list gets less airtime. A single adult with no dependants, no co-signed debt and no estate that would need cash to settle is insuring nothing. A household whose assets already cover the shortfall has self-insured, which is the same test run on a bigger balance sheet. Cover on a child's life replaces no income, because a child earns none; what it mainly sells is an option to buy cover later without new medical questions, which is worth reading out of the contract rather than assuming.

Two mechanisms matter more than they sound. Whether a survivor inherits a debt turns mostly on whether they signed for it, and in the United States on state law as well, since community property rules can reach a spouse who never signed. Federal student loans there are discharged when the borrower dies; private loans follow the lender's own terms, many of which now do the same, and federal law has required release of a cosigner on the borrower's death for private student loans taken out since November 2018. The second mechanism is that a policy pays whoever is named on it as beneficiary and a will does not override that designation, so a stale one sends the money exactly where it says. Divorce statutes and court orders are the main things that cut across it in the United States, which is a reason to read the form rather than trust the rule.

Cover through an employer is cheap and real, often a multiple of salary. It also ends when the job does, which is a poor moment to find out it was the whole plan.

Term insurance covers a stated number of years. If death falls inside the term the policy pays the face amount; if it does not, the cover simply ends. Permanent insurance, sold as whole life, universal life and their variants, covers the whole of life and builds a cash value inside the contract.

The price gap between them is structural rather than a markup. Most term policies never pay a claim, because most people outlive a twenty or thirty year term, so the pool only funds the claims that arrive during working ages. A permanent policy kept in force is certain to pay eventually, and the premium funds decades of reserves and expenses too. That condition does real work, because the premium is high and it runs for life: a permanent policy surrendered or lapsed part way through pays no death benefit at all, which is how a contract sold on certainty ends up paying nothing. For the same death benefit at the same age, permanent cover usually costs several times what level term costs, and at younger ages often more than ten times.

Term is the default because the need is usually temporary. It ends when the children are independent, the mortgage is gone and the retirement assets exist. Cover priced for exactly those years buys the most protection per dollar, which is what most households are short of.

Permanent cover has real uses: a dependant who will need support for life, cash for an estate holding an asset that cannot be split, and business agreements that must be funded whenever a death happens rather than only before a certain birthday. What it should not be is an accident, because buying it bundles two decisions, protection and saving, that are easier to judge apart.

Two mechanics get misread. Where the death benefit is level, the cash value is not paid on top of it: it is what the contract is worth if you surrender instead, and an unpaid policy loan comes off the payout. And a conversion option lets a term policy become permanent with no new medical underwriting, which is the feature that earns its keep if your health changes mid-term.

People insure the event they can picture. The actuarial tables point somewhere else. In the United States, the Social Security Administration reissues disability and death probability tables every year, and the estimate usually quoted from them is that roughly one in four of today's twenty year olds will be disabled long enough to stop work before reaching retirement age, with wider definitions putting it nearer one in three. Treat that as an order of magnitude rather than a constant, because what counts as a disability, and for how long, moves the answer more than any other input, and each new edition of the tables moves it again. What survives every definition is the ranking: through the working years, income interrupted by illness or injury is likelier than income ended by death.

Four details make the exposure larger than it feels.

Most long-term claims come from illness rather than accident. Musculoskeletal conditions, cancer, heart and circulatory disease and mental health conditions dominate the claims file. The mental picture is a fall from a ladder; the paperwork is mostly a diagnosis.

Workers' compensation, where it exists, pays only for injury and illness arising out of the job, and most disabling conditions do not. Someone covered at work can be uncovered for the thing that actually happens.

Sick pay and short-term cover run for weeks or months. The case that ruins a household runs for years, and that is the one covered least often.

Government provision is a floor, not a replacement. In the United States, Social Security Disability Insurance applies a strict any-occupation test: the impairment has to stop substantial work and be expected to last at least a year or end in death, a waiting period runs before anything is paid, and most first applications are refused. Only a minority of states operate short-term programmes of their own, so what sits on top of an employer plan depends on where you live.

Cover follows the opposite pattern to the risk. In United States benefits surveys, life insurance through work is the more widely held of the two, long-term disability cover reaches a minority of private sector workers, and individually owned policies are rarer still.

A disability policy is one of the few consumer contracts where the definitions section is the product. Two policies quoting the same replacement percentage can behave completely differently on the day a claim goes in.

Group cover through an employer often starts on an own-occupation definition and switches to any occupation after a set period, commonly two years. That switch sits exactly where a long claim turns into a lasting one, which is how a plan that paid for a year can stop.

The elimination period is a deductible measured in time. Extending it cuts the premium for the same reason a larger deductible does, by taking the frequent short claims off the insurer's book, and it only works if the household can cover the gap. The gap is longer than the number in the contract, because benefits are normally paid monthly in arrears, so the first payment arrives a month after the waiting ends rather than on the day it does. That is the job an emergency fund does, and sizing the gap is the same exercise as sizing the fund.

The cap near 60 percent looks stingy until you follow the tax. In the United States, benefits from a policy whose premiums were paid with after-tax money are generally received free of income tax, while benefits from employer-paid cover are generally taxable to the employee. The same headline percentage therefore replaces different shares of take-home pay depending on who paid the premium, so read that before deciding the number is too low.

Two methods are in common use and they are not equally good. A multiple of income, usually quoted as ten to twelve times gross earnings, is quick and has the accuracy of anything that ignores your obligations. A needs calculation asks what the death would actually cost:

• debts that would have to be cleared, the mortgage usually being the largest,
• the income the household would lose, for the number of years it needs replacing,
• one-off costs such as funeral, probate and education,
• minus what already exists: savings, investments, cover through work, and any survivor benefits, which in the United States can include Social Security payments to a child and to the parent caring for one where the eligibility rules are met.

The gap left over is the cover. Two steps get misread constantly. The first is double counting the mortgage: if it is cleared out of the lump sum, the income the survivors still need no longer has to carry the mortgage payment, and adding both at full size buys cover for the same obligation twice. The second is that a lump sum is not a salary. It has to be invested and drawn down, so the multiple needed depends on how many years the income must last and what it earns after inflation:

$$PV = \frac{1 - (1 + r)^{-n}}{r}$$

Here $n$ is the number of years of income and $r$ the return after inflation. The formula needs a positive $r$; at a zero real return the factor is simply $n$, which is what the first column shows. Read the cells as multiples of one year of income: replacing twenty years of it takes 16.35 times a year's income at a 2 percent real return. Because the rate is a real one, the figures already assume the amount drawn each year keeps pace with inflation. They also assume a year's income is drawn at the end of each year, so drawing it at the start needs a little more, they ignore tax on the returns, and they spend the balance to zero.

One thing the table cannot show is that a real portfolio does not deliver its average return in a straight line. A poor run early in a long drawdown empties the pot faster than the average implies, which is sequence of returns risk. Each factor is an expected value rather than a floor, and the further right the column, the more of the answer rests on a return nobody is promised. This is time value of money pointed at a household, and the net present value calculator runs it over uneven cash flows.

Two adjustments finish the job. A household that lost an earner lost a spender too, so replacing gross salary overshoots. And the need shrinks, which is why the term is better matched to the year the obligation ends than to a round number, and why some households hold two policies of different lengths so the cover steps down as the need does.

Disability sizing runs from the other end: start from essential outgoings rather than gross pay, check what the offsets leave, and check the elimination period against available cash. This page is educational material rather than financial advice, and what a household should hold depends on its dependants, its assets and the rules where it lives.