[int3]

> People can develop allergies during their lifetime, it's not just something that you're born with.

Sure, but the likelihood of that is low enough that insuring against that isn't worth it for most people

> But my message was about the general principle of insurance being the very thing your comment was against. The situation where the vast majority of people pay some cost of which only a few need to utilise.

No, the principle of insurance is that people pay to hedge against some event that has a reasonable likelihood of happening to them at some point. Whether it's a majority or a minority paying for it is not central to the concept of insurance.

[daemin]

Insurance only works if the majority pay but only the minority get paid. Anything else and you risk not getting paid if the event occurs.

[int3]

Nope. Assume you have insurance with a one-time fee of $X. It insures against a single kind of event that happens with 60% probability, and pays out $1.6X. So the majority of people are likely to have the event happen to them. The EV of the cost to the insurer is then $1.6X * 0.6 = $0.96X, so the books are expected to balance.

Obviously I'm ignoring variance in this calculation, but you can easily adjust the numbers to give a margin of safety.