From symbols on a page to products on a shelf
In the contingencies rung you learned to price a death benefit: A_x is the value today of a benefit of 1 paid whenever an insured aged *x* dies, and A^1_(x:n) is the cheaper version that only pays if death lands inside the next *n* years. Those were *functions* — clean machines that turn a time value of money and a mortality table into a number. This rung does something different: it takes those numbers off the page and asks what an actual person can walk into an office and buy. The two oldest, simplest answers are term life and whole life, and almost every modern product is a descendant of one or both.
The cleanest way to hold the difference in your head is a single question: *is the benefit certain to be paid one day, or not?* Whole life covers you for your entire life, so a claim is certain to arrive eventually — the only uncertainty is timing. Term covers you for a fixed window — twenty years, say — and if you outlive it, the policy expires worth nothing and no benefit is ever paid. That one structural choice cascades into everything else: the price, whether the policy builds savings, how flexible it is, and who it is right for.
Term life: pure, temporary protection
Term life insurance is the bare idea of protection with nothing bolted on. You pick a length and a benefit — "$500,000 of cover for 20 years" — and you pay a modest premium. If you die inside the term, your family receives the benefit. If you are alive at the end, the contract simply stops; you walk away with nothing back, and that is by design, not a defect. Because most people buying a 20-year term at age 35 will *not* die before 55, the insurer pays out on only a fraction of policies, so each policy can be cheap. You are buying a slice of the pool's risk, not a savings account.
Under the hood, the price of term insurance *is* the term function A^1_(x:n) you already met, scaled up by the benefit and then converted from a single lump-sum value into a stream of level annual premiums. The reason it feels cheap is honest and structural: it ignores all the late deaths. Whole life must eventually pay everyone; term quietly drops cover exactly when mortality starts to bite — in your sixties, seventies, eighties — so it sidesteps the most expensive claims. That is also its limitation. Term is brilliant at the job it claims and useless past its expiry.
Whole life: lifelong cover that stores cash
Whole life insurance keeps you covered until you die, whenever that is — so the benefit is certain, and the price reflects A_x rather than A^1_(x:n). If the insurer charged you the *true* rising cost of insuring you at each age, premiums would start tiny and balloon to thousands a year in old age, exactly when income falls. Whole life instead charges a *level* premium for life: you overpay in the early years, when your real mortality cost is low, and underpay later. Those early overpayments do not vanish — they accumulate, with interest, inside the policy.
That accumulating bucket is the policy's cash value, and it is the single feature that makes whole life feel like part savings plan. As a living policyholder you can borrow against it, or you can surrender the policy and take the cash surrender value in hand. The cash value and policy loans mechanism is genuinely useful — but here lies the field's most important honesty: that cash value is *not* a free bonus sitting on top of your protection. It is the prepaid, accumulated portion of your own future premiums. It is also almost exactly the policy reserve the insurer must hold to back its promise — money it is legally obliged to set aside, not idle cash it generously shares.
Why a level premium overpays early and underpays late (illustrative whole-life sketch, benefit = 100,000) age true cost of 1 yr's cover level premium charged difference 35 ~120 ~1,400 +1,280 overpay -> builds cash value 55 ~900 ~1,400 +500 75 ~5,200 ~1,400 -3,800 underpay <- cash value funds it 90 ~18,000 ~1,400 -16,600 The early surplus, accumulated with interest, is the cash value / reserve that quietly pays for the expensive late years.
Endowment: the savings-and-protection hybrid
Sitting between the two pillars is the endowment, written as endowment assurance. It is term insurance with a survival reward stapled on: it pays the benefit if you die within *n* years, *and* pays the same sum if you are still alive at the end. Because it pays out either way, an endowment is far more expensive than term — you are funding a guaranteed maturity payout on top of the death cover, which is why much of its premium is really forced saving. Historically it was sold as a tidy way to insure your family *and* build a lump sum for a known goal: a child's university, a mortgage payoff, retirement.
The honest caution is that bundling protection and saving into one wrapper hides how each part performs. The protection inside an endowment is no better than cheap term; the savings inside it earn whatever the insurer credits after costs, which may trail what a plain investment would return. A classic piece of advice — "buy term and invest the difference" — is really a bet that you can keep the two jobs separate and do each more cheaply. Sometimes that wins; sometimes the discipline and guarantees of a bundled product win for a real human who would otherwise never save. Neither is universally right, and a good actuary refuses to pretend otherwise.
Who each suits, and the trade-offs that follow
Match the design to the *shape of the need*, not to the sales pitch. A young parent with a mortgage and small children has a need that is large but temporary — it shrinks as the loan is paid and the children grow up — so cheap term over twenty or thirty years fits beautifully. A need that never goes away, such as leaving a guaranteed legacy, funding estate taxes due on death, or covering a dependent with a lifelong disability, calls for permanent cover, where whole life earns its higher price. Endowment fits a specific dated savings goal that you also want insured along the way.
The trade-offs run along three axes. *Cost*: term is cheapest per dollar of cover, whole life costs several times more for the same death benefit, endowment most of all. *Flexibility*: term is rigid but disposable — drop it any time with no loss because there is nothing to lose; whole life is a long commitment whose value is punished badly if you quit early, since surrender charges and the slow build of cash value mean early years return little. *Behaviour*: this is where designs quietly leak. If a policyholder stops paying — a lapse — a term holder simply loses future cover, but a whole life holder who lapses in year three can forfeit most of what they paid. Pricing actuaries must *assume* a lapse rate, and if real people lapse differently than assumed, the product's profitability swings.
Keep one limit in view as you go on. Every premium and cash value here rests on assumptions — a mortality table, an interest rate, an expense load, a lapse rate — each of which is a forecast of the future built from the past. Whole life locks those forecasts in for decades, which is exactly why it is followed historically by the more flexible universal life, the subject of the next guide, which unbundles the pieces and lets the policyholder dial them. The two pillars are not the whole skyline; they are the load-bearing columns everything else is built upon.