Most insurance is not a great bet in expected-return terms. That is why self-insuring can look so appealing on paper. If you keep the premium money, invest it, and absorb losses yourself, you may well come out ahead over time. But that is not the whole story, because the pain of a major uninsured loss is usually felt more intensely than the pleasure of having saved the premiums.
That leads to a practical rule right away: self-insure annoyances; insure disruptions. If a loss would be manageable, emotionally tolerable, and unlikely to trigger bigger problems, self-insuring often makes sense. If a loss would be hard to absorb, mentally consuming, or likely to spill into other problems, insurance becomes much easier to justify.
The basic point is simple enough that it almost sounds obvious once you say it plainly: your nervous system does not keep books the way an accountant does.
The Quick Practical Takeaway
This is why some kinds of insurance make immediate sense while others feel like nonsense in a little plastic wrapper. Health insurance, auto liability insurance, and homeowners or renters insurance often protect against losses that can hijack your attention and destabilize your life. By contrast, tiny warranties and checkout-counter protection plans often insure inconveniences more than real suffering, which is why they are usually much weaker purchases.
A broken toaster is annoying. It is not, in most cases, an existential event. If your financial plan needs an extended warranty to survive breakfast, something else has gone wrong.
Why Self-Insuring Looks So Smart on Paper
The standard objection is obvious and real. Insurance companies have to collect more than they pay out, or they do not stay in business. So for the customer, many forms of insurance have a negative expected return. That is not a scandal. It is simply the business model. And it is exactly why self-insurance can seem like the smart, disciplined, financially mature move.
Expected return: The average outcome you would expect over many repetitions of the same risk. In insurance, the buyer’s expected return is usually negative, because the insurer must collect more in premiums than it pays out in claims.
And to be fair, this is not some crank complaint from a guy yelling at the windshield. It is a real point. If the insurer did not build in overhead, profit, and a margin for uncertainty, it would not be an insurer for very long.
Prospect Theory and the Sting of Loss
The trouble is that expected return is not the only thing human beings care about, and it is often not the main thing they feel when something bad actually happens. In their 1979 paper introducing prospect theory, Daniel Kahneman and Amos Tversky argued that people do not experience outcomes as cold changes in total wealth. They experience them instead as gains and losses relative to a reference point, and losses weigh more heavily than comparable gains.
In plain English, saving premiums feels mild, gradual, and forgettable. An uninsured loss feels sharp, vivid, and immediate. The spreadsheet may treat those as offsetting dollar flows, but your nervous system does not. If you save \$1,200 a year by self-insuring, that may barely register from month to month. If you then get hit with a \$6,000 loss at exactly the wrong moment, that event does not feel like a neat balancing entry in the ledger. It feels like a blow.
That asymmetry is the heart of the case. It means a person can fully understand that self-insurance may win on average and still rationally prefer insurance in some parts of life. The point is not that people are bad at math. The point is that the math is describing one thing, while the lived experience of loss is describing another. If you are trying to build a calm life, the intensity of losses matters, not just their average cost.
Or more bluntly: saving money feels nice, but getting walloped does not feel like a fascinating behavioral economics lesson in the moment.
What Insurance Is Really Buying
This is where the Epicurean framing comes in. The goal is not to maximize the drama of winning or to squeeze every last basis point of expected value out of every decision. The goal is to reduce unnecessary pain, fear, and disturbance over the course of a life. Insurance can fit that goal even when it is, strictly speaking, a losing bet.
You are not necessarily buying profit. You may be buying stability, sleep, and the ability to recover from bad luck without your whole inner weather turning ugly. That is a much better description of what many people are actually trying to do. They are not entering a financial trivia contest. They are trying to avoid having one ugly surprise turn the whole month sour.
What Belongs on Each Side of the Line
Once you see insurance this way, the practical rule becomes easier to apply. Small losses that are annoying but contained are often good candidates for self-insurance. Replacing a toaster, fixing a cheap appliance, or covering a modest repair out of cash reserves is usually not the sort of event that deserves a stream of premiums. Many add-on protections are really selling emotional theater around losses you can handle just fine.
But when the possible loss is large enough to dominate your mind, disrupt your routines, or damage your financial footing, the case changes. Health insurance is the clearest example. Its value is not just that it may pay a bill. Its value is that it helps prevent illness from becoming financial chaos on top of physical suffering. Auto liability insurance is another strong case, because one bad event can create a mess far larger than most people should calmly agree to carry alone. Homeowners or renters insurance belongs here too, not because replacing stuff is fun, but because displacement and disruption can spread stress through every part of life.
A scuffed appliance is one thing. A hospital bill, lawsuit, or half-your-house-is-now-in-the-yard situation is another.
When Bad Luck Starts Traveling in Packs
A single uninsured hit may be tolerable. The real misery often comes when losses stack: a car problem during a medical issue, a deductible during a job interruption, storm damage right when markets are down, or some other rude cluster of events. This is where the arithmetic case for self-insuring starts to feel less triumphant.
Sequence risk makes the earlier point worse, not different. If one loss hurts more than an equivalent saving helps, then several losses in close succession can feel vastly worse than the saved premiums ever felt good. Life has a nasty habit of not spacing out its nonsense in a considerate manner.
How One Loss Turns Into Several Problems
Once that happens, the damage is not just the bill. It may mean draining reserves, carrying a balance on bad terms, postponing repairs, selling assets at a bad time, or just spending the next month with your mind tied in knots. Insurance can interrupt that chain. It cannot remove all pain, but it can stop a sharp loss from expanding into a broader season of distress. Insurance is not a perfect solution. Dealing with insurance companies can be its own special form of aggravation. But when the alternative is absorbing the full force of a major loss yourself, it is often still the least bad option.
This is one of those cases where money is not just money. It is also time, attention, sleep, and the general ability to act like a civilized person instead of a raccoon in a kitchen fire.
The Rule to Keep
So the case for insurance is not that it magically beats expected value. Usually it does not. The case is that human beings are not indifferent between a slow trickle of saved premiums and a concentrated shock. Losses hit harder than savings soothe. That is not a flaw in human reasoning. It is part of the reality any sane financial philosophy should take seriously.
If you want one rule to carry away, it is this: self-insure annoyances; insure disruptions. Keep the small stuff when the loss would be tolerable and contained. Hand off the risks that could seriously disturb your peace, your routine, or your long-term stability. Insurance is not mainly about getting paid. It is about stopping bad luck from seizing more of your life than it deserves.
That may not be the most exciting use of money. It is, however, one of the more civilized ones.