[abakker]

Devils advocate: since insurance is a business of pricing risk, isn't the point of it to use statistics to charge more to higher risk people? I've been in that group/situation before and it sucks to pay more, but, in a sense it is more fair if you pay more to be a higher risk, otherwise you end up charging lower risk people the difference.

[temp-dude-87844]

The point of insurance is to group recipients with different likelihood of risks into the same risk pool, such that it's very unlikely that many of your insured will need a payout at the same time. How you set pricing is largely a different matter.

Now, as an insurer, you have a monetary interest in ending up with as few riskier clients as you can. You can try to use pricing or exclusions to try to achieve this, but regulations often step in, because otherwise only the least risky clients could obtain insurance, and everyone else would have to fend for themselves. However, the least risky people often don't even want to purchase insurance, because they perceive its utility to be low, and the incidence of catastrophic events to be unlikely.

This generally means that clients who perceive themselves to be low-risk will place a lower value on insurance, necessitating a lower price, or they'll take their business elsewhere or nowhere at all. People who perceive themselves to be higher-risk will "accept" a higher price, because they expect to have a better outcome with insurance than without. Since companies want to avoid this type of customer, prices can creep higher than they need to be just by factoring in their risk.

[lotsofpulp]

This discrepancy happens because insurance works well for unknown risks that are independent from the entity being insured and for uncorrelated events. Similar to healthcare, vehicle collisions are highly dependent on the driver, and so some of the risk is not unknown, and hence can be priced.

Pretending you don't know the risk factor when you do (perhaps in the name of fairness) can also cause problems.

[aianus]

The low risk group (older, married, men and women) are the best positioned to absorb a rise in premiums so I'm not sure what problems would be caused.

[aianus]

Yeah but I'm sure the banks that engaged in redlining against African American neighborhoods had some stats on property values and default rates that "justified" it as well.

I don't think they should be allowed to use traits that you can't change like age, sex, and race in the same fashion as employment where you're not allowed to discriminate against women of child bearing age even if it would be economical.