[bee_rider]

I guess they aren’t very widespread anymore, but should this cover police who ride motorbikes? Or farm/ranch workers (they might ride ATVs)?

I guess we could do something like:

    <normal coverage> - <adjustment for risky behavior> + <adjustment for pro-social outcomes> 

But I think we will have trouble puzzling out the last term!

[lostmsu]

One has to draw the line somewhere. What you are doing is called a slippery slope fallacy.

[bee_rider]

I’m not sure it is a slippery slope. With a slippery slope we expand the scenario through a sequence of “if X, when what’s to stop Y,” right?

Motorcycle cops are an obvious subset of people who ride motorcycles. It isn’t an extension at all to include them in your logic.

ATVs might be more of an extension. But, I bet if we wanted to we could find all sorts of jobs that are more dangerous than motorcycle riding.

(Edit: just to be specific, you say we have to draw the line somewhere. Well, then where?)

[lostmsu]

There's a long list of topics where this particular reasoning could draw a line somewhere. It is unfeasible and pointless to cover them all unless they are all banned or all allowed (this essentially is the current state +- AFAIK).

I'd say it is worth looking at redrawing that based on the maximum effect achieved. Drugs would be at the top of this list, followed by motor vehicle use and unhealthy foods. There is probably not enough justification to go beyond the 3.

[bee_rider]

I’m not clear on what the effect actually is. If it is cost reduction, not sure where motorcycles should be on the list (they are probably more costly for life insurance agencies than for health insurance ones…).

I guess I’ve been beating around the bush, but my point is that targeting drugs specifically for this sort of thing would seem kind of, I dunno, puritanical to me (as someone who doesn’t partake). I’d rather just insure everybody and hope they don’t hurt themselves, just out of their own self interest.

[lostmsu]

Puritanical is just a label here that you slap on the idea you don't like. In the US drug overdoses are the top cause of death under 35. I have no idea how to properly estimate severe harm that does not result in death but if you take say war as the first approximation, you can 4x the deaths to get non-fatal severe cases.