I would be surprised if the stimulating effects outstripped the dying early effects, given that in one case you are less performant and in the other you cease to exist.

I agree that we should be honest with the stats, which is why I think it was disingenuous to claim that $500 billion was wildly off when the reported costs from the CDC were a substantial portion of that. Sure, we can start considering secondary effects like cost savings from people dying early and productivity lost, etc. etc. But all I'm saying is that I don't think the original claim was on its face absurd.

Agreed that we shouldn't be doing some absurd cost-benefit analysis for the economy when considering whether to combat smoking, obviously people dying is sort of a priori bad.

That’s reasonable. Personally I doubt the stimulant effects come close to zeroing out the lost productivity from heath issues and early deaths. But, at these scales it’s a multi billion dollar effect and needs to be accounted for.

Also, if you object to wildly off base that’s fair. I would call even the 3x difference between the CDC’s 170B and my simplified 500B sanity check as wildly off base. But, that’s just my scale +/- 50% is fine, 2x raises read flags for me, but out at 3+x is a fundamental incompatibility in different estimates.