[perl4ever]

I feel like there is an obvious reductio ad absurdam. If you are obese, yes, you can expect higher costs. If you are diagnosed with cancer, you can expect higher costs. If you are on your way to an appointment to check for cancer, you are higher risk than someone who has no cause to worry. If we take the idea of discriminating against known risk to an extreme, there would never be a payout at all, because hours, minutes, seconds, or milliseconds before a claim, there is always an indication of elevated risk. If you have full telemetry from all the cars you provide collision insurance for, you can determine that someone is an unacceptable risk moments before impact. "High frequency insurance underwriting" would only be a difference in degree from the sort of discrimination people defend, but it would make the entire industry pointless.

Therefore, I believe that everybody agrees discrimination against known risks must be limited by regulation or other means to make insurance viable; we're just arguing about where to set the boundary. It is generally accepted that race, religion, and so on are prohibited forms of discrimination, but that doesn't mean there is some principle that says all other forms of discrimination are ok. Furthermore, the arguments for discriminating against the obese logically apply to protected categories, so how do you reconcile that?

Apart from logic, I don't understand why anyone would want to argue in favor of discrimination against people who have a tendency to be obese, even if you don't have that particular issue. There are an unlimited number of conditions a person could have that could be similarly discriminated against.

[conanbatt]

You are describing a world without insurance: where each incident bears its full results on the bearer. This is the situation with almost anything you do everyday: from eating to crossing the street, to wearing helmets on electric scooters.

Insurance as a product is a way to manage that risk, but it doesnt eliminate it, and as a product it makes the total cost higher, as you need to pay to manage the risk. Insurance itself always bears the risk of adverse selections: it attracts those that will have higher tendencies to abuse the insurance.

That is the nature of insurance, what it solves and why it requires human effort to manage, it can't be automated away: it is an economic solution.

Another problem would be: some people have very expensive medical treatment and we should not as a society let them die or suffer death from it. But that problem is not soluble through insurance, that is soluble through charity and charity alone. Using insurance as a mechanism of charity is using a hammer on a screw.

[perl4ever]

Yes, I was describing a world without insurance. I'm saying a world with insurance and without any of what you call charity is strictly speaking impossible. And therefore the separation between insurance and what you call charity is spurious.

Insurance cannot exist and solve any problem if insurers are able to discriminate perfectly, so it cannot be a matter of principle that they should be able to discriminate to any arbitrary degree.

The idea that insurers are inherently justified in discriminating against poor risks is a nihilistic attack on the idea of insurance.

Total information awareness would lead to perfect discrimination and the annihilation of insurance, so you cannot justify by a principled argument discrimination against those you don't like purely on the basis that insuring them is charity.

[conanbatt]

> I'm saying a world with insurance and without any of what you call charity is strictly speaking impossible

Truly don't get this. Insurance discriminates before the fact: it's a framework to gauge risk, and collect a premium for managing it. There is no charity involved in that process.