Surely most living people are worth more than the economic cost of keeping them alive. They don't seem to even mention the economic value of a living person, other than to say that they didn't take into account the difference in productivity between obese and non-obese people.
Given that most smokers/obese people tend to die right around retirement age, we probably aren't leaving much on the table (from a purely economic perspective). Smokers probably aren't missing enough days of work to cancel out the $60k they save the system by smoking (not to mention all the extra taxes they pay).
But there's more value to human beings than their economic value, hence my talk about non-monetary costs. There's very real suffering that happens, and I'm not sure that's something you can put down in a spreadsheet somewhere.
Lots of tremendously valuable labor is done by people who don't pull down a wage. I.e., retired people who volunteer and watch the grand kids. Many people -- perhaps most -- contribute more value to society in retirement than in their wage labor years.
I mostly just disagree with the framing of the question. But, yes, I reckon the average retired person contributes as much/more to the GDP of the country than the typical wage-earning person.
There's no reasonable way to choose between the two framings. They are both obviously important. An accurate and purely quantitative utilitarian conversion of human suffering into USD is obviously beyond any reasonable assumption on the bounds of human rationality. A purely non-economic assessment ignores the possibility of real zero-sum tradeoffs caused by hard bounds on productive capacity.
Almost all economic activity is not zero-sum. Contrary to the intuitions of household budgets, at the scale of an economy operating over the timeframe of a generation, spending a dollar on X probably doesn't actually mean taking a dollar away from Y, especially if the productive/extractive inputs to X and Y are very different.
Therefore, it's more about finding a (necessarily vague & qualitative) exchange rate between the two. But at the order of $60K/yr in costs, I think that conversation is moot. Let's say retirees produce half of that value -- a grossly conservative estimate. That means we value "all the rest" at $30K/yr, which is prime facie absurdly low (what does a new SUV cost these days? How many "SUV new car feel vs. Corolla new car feel" units of happines equals the units of unhappiness caused by lung cancer? You get the point.).
If the number were $6M or even possibly $600K then the resource allocation question becomes more reasonable because there we approach the realm of starting to consider real zero-sum tradeoffs in what is possible. But at $60K with 50% to >100% offset, the number becomes too small to worry about the exchange rate. At most, in a pathologically broken system, we might, possibly, be discussing what we choose to do (not what is actually possible at minimal sacrifice). But it's not even clear to me that this is true in the US!
So, the framing seems silly, because once you do the work to figure out how to properly compare two radically different things you arrive at the conclusion that $60K is an order of magnitude away from forcing any sort of actual zero-sum resource allocation decisions.
