3 ms·

> Are they wrong? Did you not just increase their liability by 0.001%? If people with lower risks didn't pay for people with higher risks, there'd be no insura

> Are they wrong? Did you not just increase their liability by 0.001%? If people with lower risks didn't pay for people with higher risks, there'd be no insurance market. If you can perfectly predict what's going to happen to someone - and you charge them all exactly what it's going to cost, + your overheads, then there's no reason they should buy insurance rather than putting the money into an investment plan. Actually they'd probably be better off with the investment plan. Insurance is like gambling, what does the insurance company know that you don't? What do you know that they don't? If they get really good at predicting your health, then being accepted for an insurance scheme would be a really strong argument to do something else with your money.

diminoten 13y ago
It's a spectrum, not a boolean value. Insurance companies will never get to 100% omniscience, that's silly. Nor will lower-risk people pay so little as to not offer an incentive for insurance to exist. Lower risk is not no risk. Today the concept I'm talking about already exists in the form of good driver incentives, so the outcome you're predicting is already not how this plays out.
- 6d0debc071 13y ago
  A lower risk insurer can afford to offer lower prices. Similar things happen in the credit market, where risk is more easily predicted. High risk credit is very expensive, often pricing people clean out of the market - door to door credit is declining in popularity as credit agencies find their margins getting slimmer. Even today, some people are to all intents and purposes uninsurable. Perfection isn't required. Just knowledge of a significant disparity.
enoch_r 13y ago
This is simply not true. People are risk averse, and buy insurance to protect them from unlikely negative outcomes, not because they believe that the insurance company will, on average, pay out more than they pay in premiums. I can invest $200 a month for the next 10 years, or I can buy health insurance. On average, I will have more money after 10 years of investing. But the additional $24,000 of savings will increase my happiness very little relative to getting the healthcare I need in the case of a catastrophic injury or illness. So I'll gladly lose a little bit of money every month (statistically) to avoid the possibility of that outcome. Edited to add: one way to think of this is to pose the following thought experiment. Imagine that tomorrow you have a 50% chance of doubling your wealth and income, and a 50% chance of losing everything. What would you be willing to pay to guarantee that you keep your current wealth and income? If you're like most people, the answer is nonzero and positive. That's how insurance companies make money.
- 6d0debc071 13y ago
  This is a good point, and I have to admit I hadn't thought of it from quite that angle. I guess I'd have to reiterate my point to diminoten; that being in the same risk pool as people with a higher risk is a questionable decision. Certainly my risk might not be perfectly predictable, but to the extent that it is predictable, I don't see why I should pay for you as well. If I set up a company that just took enough from you to cover your predicted risk - and the risks of people like you - plus some small charge for me, I could charge you less for it than a company that tried to cover everyone.
maxerickson 13y ago
You are conflating risk and outcome. All drivers are at some risk, some bad and lucky drivers will never have a bad outcome, some good drivers will have multiple bad outcomes. Risk of injury is similar. I would expect the understanding necessary to accurately predict cancer to be very useful in treating it (it should also have much potential to reduce costs...).

diminoten 13y ago

It's a spectrum, not a boolean value. Insurance companies will never get to 100% omniscience, that's silly. Nor will lower-risk people pay so little as to not offer an incentive for insurance to exist. Lower risk is not no risk. Today the concept I'm talking about already exists in the form of good driver incentives, so the outcome you're predicting is already not how this plays out.

6d0debc071 13y ago
A lower risk insurer can afford to offer lower prices. Similar things happen in the credit market, where risk is more easily predicted. High risk credit is very expensive, often pricing people clean out of the market - door to door credit is declining in popularity as credit agencies find their margins getting slimmer. Even today, some people are to all intents and purposes uninsurable. Perfection isn't required. Just knowledge of a significant disparity.

enoch_r 13y ago

This is simply not true. People are risk averse, and buy insurance to protect them from unlikely negative outcomes, not because they believe that the insurance company will, on average, pay out more than they pay in premiums. I can invest $200 a month for the next 10 years, or I can buy health insurance. On average, I will have more money after 10 years of investing. But the additional $24,000 of savings will increase my happiness very little relative to getting the healthcare I need in the case of a catastrophic injury or illness. So I'll gladly lose a little bit of money every month (statistically) to avoid the possibility of that outcome. Edited to add: one way to think of this is to pose the following thought experiment. Imagine that tomorrow you have a 50% chance of doubling your wealth and income, and a 50% chance of losing everything. What would you be willing to pay to guarantee that you keep your current wealth and income? If you're like most people, the answer is nonzero and positive. That's how insurance companies make money.

6d0debc071 13y ago
This is a good point, and I have to admit I hadn't thought of it from quite that angle. I guess I'd have to reiterate my point to diminoten; that being in the same risk pool as people with a higher risk is a questionable decision. Certainly my risk might not be perfectly predictable, but to the extent that it is predictable, I don't see why I should pay for you as well. If I set up a company that just took enough from you to cover your predicted risk - and the risks of people like you - plus some small charge for me, I could charge you less for it than a company that tried to cover everyone.

maxerickson 13y ago

You are conflating risk and outcome. All drivers are at some risk, some bad and lucky drivers will never have a bad outcome, some good drivers will have multiple bad outcomes. Risk of injury is similar. I would expect the understanding necessary to accurately predict cancer to be very useful in treating it (it should also have much potential to reduce costs...).