Suppose spouse1 and spouse2, aged 55, and children aged 17 and 19. Both spouses purchased 20-year term life insurance 19 years ago. The insurance is no longer needed to support either spouse or children in the event of a spouses's death.

The bill for the 20th year has arrived. The question is whether to pay it. This looks like a lottery ticket question: pay a small amount, usually win nothing, and very occasionally win big. In this particular case, the premium is $175 and the payout is $250,000. $175/$250,000 = 0.07%. So if a spouse's probability of dying in the next year is greater than 0.07% then it probably makes sense to pay the premium for that spouse. One can use data from the CDC to estimate the probability of death and thereby make some kind of reasoned choice about whether to pay the final year's premium.

It's usually the times when my logic feels airtight that I am mistaken. Is my analysis correct?

  • 5
    Mortgage or rent? Do both spouses work? Healthy (no pun intended) savings and retirement fund? What if both spouses die?
    – RonJohn
    Commented Nov 30, 2019 at 16:57
  • 33
    Your analysis should also take into account the relations between the spouses, and whether either has access to rare, undetectable South American poisons :-)
    – jamesqf
    Commented Nov 30, 2019 at 18:03
  • 2
    I'm in much the same situation. I'm 69 in 18 days. Life insurance paid yearly ends at 70 max. I have health issues which COULD kill me at any moment but PROBABLY wont. Do tell me what you decide :-). Commented Dec 2, 2019 at 1:03
  • 2
    @RussellMcMahon COULD and PROBABLY are relative. Any individual COULD die at any given moment, but PROBABLY won't.
    – glglgl
    Commented Dec 2, 2019 at 8:26
  • 2
    @glglgl My could's and probably-wont's are a bit better (or worse) correlated than that. I take heart rate control and rythym control medication and blood thinner (anti coalgulant)). Typically in the 4 to 6 week range I get bouts of fast Atrial Fibrillation with high blood pressure. Pulse can sound like minuet in G played staccato. AF usually lasts under a day. About 6 days after the AF +/- 1 day (asymmetric) I get Migraine aura alone (no pain, headache, other effects). I've never had migraines but it's a very very good match to the CSD brain depolarisation that causes migraine aura. ... Commented Dec 2, 2019 at 9:13

9 Answers 9


Term life insurance is something that you purchase when the financial loss that would be incurred by your loved ones if you died would be catastrophic. In your example, 20 years ago when you were expecting a baby, most people would be well-advised to purchase term life insurance.

Now that your kids are (nearly) adults, and the insurance money would not be needed, you wouldn't make that new purchase today. As Hart CO said in his answer, life insurance is structured to make a profit for the insurance company.

However, one thing to consider is that the term life insurance payments are structured so that the payments are uniform throughout the term, but the odds of your death go up as you get older. This means that, from an expected outcome perspective, you are overpaying for the life insurance in year one, and you are underpaying for the insurance in year 20. If being covered in the last year before your youngest turns 18 helps you sleep at night, rest assured that you are likely getting a bargain for your last term life insurance payment. If you were to renew your insurance for another term, the payment would be much higher.

  • I doubt that 'but the odds of your death go up as you get older' is true from mathematics point of view, to explain this better, among guys who live to 67 percentage of the guys who survive until 68 is pretty high as those guys lived all previous years and are probably healthy, while highest risk of not surviving is among kids. Basically if you know a person would die, yes true, more likely the person is adult, but if you claim that a (one) selected person has higher percentage of dying while aging than before, you are wrong.
    – Igor sharm
    Commented Dec 2, 2019 at 17:06
  • 11
    @Igorsharm That's the point of indemnity insurance-- you don't know a person's individual risk. But average mortality risk does tend to go up as people age, in general. The trend you describe does exist (though you may be overstating it somewhat), but the actuarial process used in designing and pricing the insurance policy almost certainly have taken it into account.
    – Upper_Case
    Commented Dec 2, 2019 at 17:27
  • 18
    @Igorsharm That's a solid theory, but we don't have to guess--real world actuaries have already spent a LOT of time figuring out this exact question. The chance that you will die in the next year decreases from birth until about age 10, then starts a steady upward trend that never reverses. For any adult, even considering that you survived all previous years since your birth, your average mortality chance does in fact steadily go up with your age. Example: ssa.gov/oact/STATS/table4c6.html Commented Dec 2, 2019 at 20:26
  • 1
    @Igorsharm Here's ages 0-99 of the data from GrandOpener's link as a graph: i.sstatic.net/BVlpa.png You can see the initial spike drop by year to age 10 or so, then climb inexorably from there. Commented Dec 3, 2019 at 13:21

The insurance is no longer needed to support either spouse or children in the event of a spouses's death.

Life insurance is pretty much only for people who aren't in your situation. You know that the insurance company is making money by using their actuaries to come up with very accurate payout probabilities for large cohorts of their customers. Since their math determines your premium and they make a profit, it's a losing proposition from your end unless you need the insurance to provide for those left behind.

You're right though, it can be viewed as a morbid lottery. If you had inside information that they were not privy to that significantly changed your odds then it could also be financially worthwhile. Though, depending on your policy not disclosing certain things could result in reduced benefit. The house always wins.

  • 12
    Also, with a fixed premium for the entire term, it's possible that the premium:payout ratio in the later years is actually in your favour, the insurance company having already made enough profit from the earlier years to cover it. Commented Nov 30, 2019 at 19:54
  • 2
    Your last paragraph reminds me of the Simpsons episode where Ned Flanders' uninsured house is destroyed in a hurricane - "Neddy doesn't believe in insurance, he views it as a form of gambling". He's not wrong. Commented Dec 2, 2019 at 15:11
  • 3
    @Nuclear Wang If insurance is a form of gambling, then what isn't? If I buy a widget with no warranty, I'm gambling on it not being DOA. If I buy a widget with a warranty, I'm gambling because the warranty is insurance. I would say that the difference between insurance and gambling is that insurance decreases the variance of your future net worth, while gambling increases it. They're alternatives, not equivalents. Flanders was gambling by not insuring his house.
    – benrg
    Commented Dec 2, 2019 at 22:58
  • 1
    Because the premium hasn't increased over the years, but the chance of your death in each year has, it's like a morbid rolled-over lottery. The payout is higher than it would be if you bought a new policy paying the same premium.
    – nigel222
    Commented Dec 3, 2019 at 16:58
  • I'd ask, what is the utility value of saving that small premium to myself? If I have adequate savings and am living within my means, it's probably small. So let it run?
    – nigel222
    Commented Dec 3, 2019 at 17:00
  1. "no longer needed to support either spouse or children" -- needed != valuable. "Needed" is ill-defined anyway -- I doubt even when you thought it was needed that your family would have starved to death without it. Would a death now still impose a financial cost, even if not a debilitating one? If so, then risk aversion still says the insurance is worth somewhat more to you than its strict expectation value.

  2. As has been noted a few times here, even on an expectation value basis, the insurance is likely now a good deal since the premium has presumably been level while the chance of death has been strongly increasing with age. If you cancel, you are throwing away the year that is most in your favor.

  3. Another consideration is whether you can truly do a penalty-free cancellation (only consequence is loss of this policy), or whether this would be considered defaulting on a contract, with additional consequences for your credit report, future dealings with this or other insurance companies, etc. Really, if you decide to cancel by withholding payment in this specific circumstance, the insurer should send you a nice thank-you gift because it's a windfall to them. But their automated customer rating system might not have enough nuance to see it that way.


I've noticed that it's a recurring theme of whether insurance company is overcharging etc. While it's easy to think it's a zero-sum game and the fact that they make money means you lose, reality is more complicated. For one, insurance companies have access to investments that individuals don't. Also, they mis-priced products all the time. (My guess is LTC gonna blow up for sure.)

Anyway to OP's question - one flaw in your logic is that both CDC data and the mortality table the actuaries use are based on large number of samples. That makes things easy to predict. Individual situation is the hardest to predict. That's why they don't do it.

If I were you I'd simply think about whether the premium is something I'm comfortable paying for the peace of mind (as opposed to trying to find out if there's an arbitrage opportunity). And if you are fine with the worst case scenario without the insurace payout, just drop it and stop worrying about it.


The problem with your logic is that both sides of the insurance contract understand that logic and the actuaries at the insurance company aren't going to let the policy get priced below its expected payout plus a healthy profit.

  • 1
    Term life is a product with a lot of competition, keeping prices reasonable. Yes, the insurance company makes a profit, but the level payment over the 20 year term ensures that the company makes their profit early in the term. If OP cancels his plan in year 19, the insurance company brings in less revenue, but also does not need to cover the OP during the riskiest portion of the term. It benefits the bottom line of the insurance company if lots of people skip their last year.
    – Ben Miller
    Commented Dec 2, 2019 at 18:09

Statistics will not give you enough information for a good decision in the situation that you describe. There are enormous differences among 55-year-olds.

Supplement the statistics with the data from a recent comprehensive physical exam to get better guidance. If you haven't had a comprehensive exam in years, buy the insurance. Add to that the information from your fathers' and mothers' health histories, how much you drink and smoke (if at all) and exercise, and whether you fasten your seat belts religiously, and check your gear when you go base-jumping. If you feel good about all that, consider spending the $175 on a luxury.


If you look at it logically, you'd ask your kids to pay for your life insurance once they can afford to pay for it.

Say I'm 19 years old, my dad is 55, and he has a $250,000 life insurance that costs $175 per year. By paying $175 for 30 years I get $250,000 when dad dies at 85, for a $5,250 pay. But look at it from dad's point of view: He isn't making any money from the life insurance. He has no reason to pay for it (if I can afford to pay for it).

Are you sure the numbers are right?

  • 6
    That's not what the numbers mean. This is the last year of a term policy that covered death from age 36 to 56. A new policy would cost much more.
    – nanoman
    Commented Dec 1, 2019 at 23:43
  • 3
    Yeah, you might want to check prices on that policy. Death is certain; they're not going to allow you that kind of win; that would be a "Free Lunch". There's no such thing. Commented Dec 2, 2019 at 3:29
  • 1
    The thing you're forgetting: no insurance company on the planet would insure an 84-year old for $250,000 for $175/year. You will not get a 30-year level term policy for a 55-year old for anywhere close to that. (As @nanoman points out)
    – spuck
    Commented Dec 2, 2019 at 18:06

Tim Harford What makes gambling wrong but insurance right? explain your dilemma in details.

Risk-sharing mutual aid societies are now among the largest and best-funded organisations on the planet - we call them "governments" <-- lengthy examples snipped> Providing insurance is no longer a mere money-spinner for governments. It is regarded as a core priority to help citizens manage some of life's biggest risks - unemployment, illness, disability and ageing.

A term life insurance is essential if there isn't sufficient social safety net for individual basic needs. If that is the case, even the odd of pass away is very odd, it is not worth to save the $175.

Additional note : In theory, if a regulated large insurance entity takes over the business of the universal risk pool than the individual private insurance company, it may reduce the term life insurance premium. Due to low-cost data collection and demography aggregation, a private insurance company will try to impose a higher premium on the high-risk group to reap more profit(insurance company may charge higher premium on an individual with a blue-collar job than one with a desk job). Sooner or later, it will transform into a similar problem like private healthcare insurance: those unlikely to use the insurance don't need it, while those who need it cannot afford it.

  • Could you clarify how this answers the question, particularly without reading the contents of the link? Commented Dec 2, 2019 at 19:04

One thing to consider is that this is a type of savings account, essentially. You've been paying into it, although you hope not to get anything out of it any time soon. Depending on your policy, once you hit a specific level of payment into the life insurance account, the interest on what you've paid in may actually pay for the premiums from there on out, so you may no longer need to pay in monthly to continue to be covered.

In this way, you can stay covered until it's needed, then you essentially get your money back. Not all insurance works this way, but I've heard of policies that do. Maybe they are more rare than I expect, but it's something to look into.

What I'm saying is that if you stop your policy now, you could be losing all the money you've paid in as well as any interest you might be earning on it in the future.

Also, you might be well off now, but what if someone loses their job, gets in an accident, becomes terminally ill, or the stock market crashes your 401k and other investments.

The older a person gets, the more likely it is to get a terminal illness. In today's age of massive medical bills that health insurance doesn't always cover, what you think of as a nice nest egg could disappear rather quickly. The life insurance can be used to pay for any "leftover" medical bills after death, so don't totally discount it's usability.

Life insurance can be looked at as a lottery ticket, but is it one that you want to ignore completely? It happens to be the only one that's almost guaranteed to be worth something.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .