Why should we consider life expectancy in the dialysis, nursing-home and hospice era to somehow be a positive measure of societal health?

A older friend of mine who structured financial products said that the worst thing we did for the healt of the Social Secuirty fund was the war on cigarettes. He said it in jest... but he wasn't wrong.

Oh, I'd love to hear more detail on this if you remember any?

The TLDR is: smoking and obesity cause you to die sooner with less total medical costs[1]. This is great for things like social security since you can pay in over your entire life but never collect.

The story I remember hearing (and couldn't find a specific citation) was that anti-smoking groups made up a bunch of big numbers about smokers' high medical costs to scaremonger. To fight back Phillip-Morris did a study and legitimately found that smokers died sooner and total costs were lower. Unfortunately, that's not something you really want to publicize, so anti-smoking people can keep making up scary numbers and cigarette companies can't really fight back.

[1] http://www.nytimes.com/2008/02/05/health/05iht-obese.1.97488...

If I remember correctly he was structuring a derivative contract based on life expectancy/actual life time. The goal was to hedge risks between health insurers and life insurance policies.

Life insurance policies policies want to offset the risk of a group dying prematurely in the defined time window and health insurance companies want to offset risk a group of people living long but with signifiant health problems.

At the time it didn't go anywhere because some folks dubbed it "Death Derivatives" and he was working on it at the mortgage crisis was unfolding.

Probably because smokers tend not to live as long, and therefore draw fewer benefits.