4 ms·

I think you're missing something important in this. If you make the fundamental assumption that what happens to people has a level of randomness to it (don't t

by robkop 8y ago

I think you're missing something important in this. If you make the fundamental assumption that what happens to people has a level of randomness to it (don't think we'll ever be able to escape this assumption) then insurance always will have a point even if your bucket size is the size of the individual. This is because insurers provide a hedge against a certain amount of negative randomness over a time peroid. They calculate the risk on a policy as if they provided it to you a million times and then base their pricing on that. Since they do this across a body of customers the pooling system still functions, it just is more accurately priced. So as long as there is a risk instability over time they eill have the ability to provide a service worth paying for.

mikeash 8y ago
You're right that there's always some utility as long as there's randomness, but that utility decreases the more accurate your predictions get. Rather than a magical death clock, let's say that actuarial science gets to the point where it can figure out an individual's life expectancy ±5 years with a 90% hit rate. The possibility of being hit by a bus or having some unknown wild-card disease means you could die much younger, and you might beat the odds and die much older, but 90% of people die within their predicted range. Now, Bob goes to sign up for life insurance and he's predicted to die at age 50. He really wants this insurance to ensure his family can survive without him, but the premiums are completely outrageous, so he passes. Alice is predicted to die at age 100 so her premiums are cheap, but she sees little point in it, so doesn't sign up. In this world, I see a lot of accidental death insurance being sold, but almost no general life insurance. Some people would buy it, as it would have some utility, but not a whole lot.
- Spooky23 8y ago
  Bob needs to participate in a group policy.

by robkop 8y ago

mikeash 8y ago

You're right that there's always some utility as long as there's randomness, but that utility decreases the more accurate your predictions get. Rather than a magical death clock, let's say that actuarial science gets to the point where it can figure out an individual's life expectancy ±5 years with a 90% hit rate. The possibility of being hit by a bus or having some unknown wild-card disease means you could die much younger, and you might beat the odds and die much older, but 90% of people die within their predicted range. Now, Bob goes to sign up for life insurance and he's predicted to die at age 50. He really wants this insurance to ensure his family can survive without him, but the premiums are completely outrageous, so he passes. Alice is predicted to die at age 100 so her premiums are cheap, but she sees little point in it, so doesn't sign up. In this world, I see a lot of accidental death insurance being sold, but almost no general life insurance. Some people would buy it, as it would have some utility, but not a whole lot.

Spooky23 8y ago
Bob needs to participate in a group policy.