Over $20.000.

I got that figure from http://content.time.com/time/health/article/0,8599,1808049,0... which suggests that a year of life should be worth about $130,000. So that is a bit over $10k per month.

Furthermore your hypothetical is slanted to minimizing the problem. Sure, these people are often old and sick. The median age of those killed is around 80. But they were Generally not on death’s door. From https://www.ssa.gov/oact/STATS/table4c6.html your average 80 year old has over 3 years to live. So we should be willing to spend at least $400k. Per life at the median. We are starting to talk about real money now.

But even so it is still low. Because the arithmetic average of this grizzly calculation is pulled up a lot by the younger people who die.

OK, I see you retort, so you’re coming to an order of a magnitude of a trillion dollars for the USA. But aren’t we doing more trillions in stimulus? Does that make sense?

The answer is yes. When we “spend” trillions in stimulus, we get most of it back. (In 2008 we actually made a profit.)

So even if you want to argue straight dollars and cents like an actuary does, we still come out ahead. Preserving human lives, even old people who do not have long to live, is valuable.

Lets say average life expectency of victims if they didn't have COVID is say 15 years.

Lets say 2 million would die without a lockdown.

That's 30 million years of life without a lockdown.

In the U.S. a year of life is under $150k, so that's an upper "value" of lockdown of $4.5T

U.S. GDP is 21T, so a 25% drop in GDP caused by a lockdown would not be value for money.

This is an objective way to look about it. One important element missing is how much we'd have to pay in economic loss if we ended the lockdown today....

Just think about it. Will life simply return to normal with everybody going to beaches, stores, parks etc if they know the risks giving the new awareness of the disease? What I'm saying is that there is a high chance of a recession from the fear alone (not saying "irrational fear", simply "fear") as people will stay home and take fewer risks as they normally would, with the associated economic damage. Many will simply refuse to go back to an office.

So when calculating the savings from ending the lockdown, you have to take in consideration our losses from people's voluntary self-quarantine until the pandemic is effectively over if ending the lockdown prematurely.

Factors we don't know include how behavior would change even without top down pressures (i.e. doing nothing will still damage economic output), and how long lockdowns can last before protests, riots, and full on civil war, how effective a lockdown would be, what pressure there will be on healthcare providers in each scenario (including say healthcare workers refusing to treat patients)

There's also the subjective view - we aren't robots, people aren't objective automatons, many of us play the lottery for example, despite on average ending up behind, because we value the idea that we could get rich more than the value of the money we'd save not playing it.

There's then the relative weighting -- who suffers from that 25% fall in GDP? We may be willing to accept a 0.1% chance that someone in our close family will die in a car this year for $50k of personal gain, but we won't accept a 50% chance for "the economy" gaining $100m, of which $99m will go to "the 1%".

A crashing GDP affects "the 1%" more in financial terms than the majority. On the other hand with no cash to pay

There's then the value of a life. To me, my life and my family's life is far more important than yours. That's human nature. In the trolley problem where the choice is between killing their child and diverting it to kill 2 anonymous KKK members, almost everyone will divert it. Will a typical 25 year old with no family, a minimum wage job, and no health insurance, be willing to lose their home and live on the streets for the next 10 years in the unlikely save an 80 year old billionaire who doesn't care about them?

Good points. There are too many variables, and unknown changes to said variables for us to predict an optimal long term policy for handling this.

Thinking of it as objectively as I can, I can't see a scenario where the country keeps functioning well with the massive, rapid number of deaths associated with Covid-19. I know people die by the thousands in car accidents and other things. But at least in the short term, the widespread panic from a novel disease in ~2x/day geometric growth with 0.5-1.0% death rate (at best) would far exceed in severity anything we've ever seen.

Imagine Italy with no lockdown, would that be conducive to a healthy, or just mildly impacted economy? I don't see it. We had time to act, we didn't, now every option is bad, and now we're feeling the pain of the "less bad" option.

That is a good comparison point, and one I hadn't thought of.

Currently Goldman Sachs is estimating that next quarter will see production drop by 24%. If production drops for a single quarter by that much then rebounds, this is still worthwhile on that estimate. If it drops for a whole year by that much, you are right that the implicit value we are giving these lives is more than we are willing to pay to save lives in other circumstances.

The idea that every one of the millions dead from COVID-19 would've died anyways in the same year is... unsupported, to say the least.

The idea that many of them would have died anyways within a year or three is rather better supported, though. As is the idea that we need to run a cost-benefit analysis here.

There are far too many unknown, and indeed unknowable variables, for a cost-benefit analysis, even if it were politically acceptable.

There are broadly three options the US has

1) Total lockdown

2) Sporadic geographic lockdowns of various intenstiy

3) Do nothing

Factors we don't know include how behavior would change even without top down pressures (i.e. doing nothing will still damage economic output), and how long lockdowns can last before protests, riots, and full on civil war, how effective a lockdown would be, what pressure there will be on healthcare providers in each scenario (including say healthcare workers refusing to treat patients)

While our higher brains knows that as a society we are unwilling to spend $1 billion to extend someone's life by 2 days, when it comes to this sort of magnitude there are other considerations than financial benefits, indeed the entire situation brings questions of what is the value of money anyway. When we say we're willing to spend $150k to extend a life by a year, what does that actually mean?

See https://news.ycombinator.com/item?id=22751269 for a cost benefit analysis.

Long story short, it is worth it.

As with iso1631 there, I disagree. The lockdown will only save significant numbers of lives if it lasts until we have a way to reduce population death rates (either via better treatment, prophylactics, or vaccines) OR if we can exterminate it entirely and prevent COVID-19 from becoming endemic to the US again. I think the latter is unlikely, and the former requires 6+months of lockdown.

If we do three months of lockdown and then people give up and head out as normal again, we've burned a few trillion dollars, and a hundred million years of US citizen lives, for nothing. The deaths still come, just shifted back three months. That's the worst case, and it's what we might be heading for.

It should be entirely possible to ramp up masks and ventilators in a couple months. China is already making immense amounts.

Universal mask wearing would go a long way, especially with a much lower baseline of active infections.