Interesting. This baselines the cost of one life to $2000 effectively. So I could trade two lives for a top of the line Macbook Pro. Buy an SF home for 500 lives. I spend one human life to live in SF every month. Huh.
If you're interested, you should read GiveWell's detailed analysis of why it's reasonable to believe this (i.e. the efficacy of the charity to date, whether additional money will be spent with similar efficacy, etc.) and really form an opinion for yourself about whether it's true. Then, act on that opinion.
Very cool. Thank you for that. I quite enjoy considering these.
Interestingly, offering myself a choice between an arbitrary human life and a nice meal, I find myself choosing the meal. So I value the arbitrary human life at under a hundred dollars.
Just curious to me. I would have expected to have picked a higher valuation.
I enjoyed that. Interesting to me that I value arbitrarily children so low. Getting my turbocharged engine vs. getting the same car without a turbo could have saved a couple of lives. I suspect I would have made the same choice and killed two kids for a turbo if I had thought explicitly in those terms then.
Kind of puts the trolley problem in a new perspective -- forget about saving five people or only one, it's more about whether or not you'd save five people or buy that long-range upgrade for your Tesla.
I get your point, but don't be too down on yourself. Nobody is running around the Against Malaria Foundation saying "quick, buy 125,000 more nets!". Rather, they take in money, and they assign something tangible (in this case, a net that's extrapolated to a life) to each donation so it feels a bit more connected.
This isn't to even remotely take away from what Derek did – I've been a fan for years, and this was one of the coolest blog posts I've read in a long time. It's the opposite... I just don't want anyone to feel like they're killing someone every time they pay rent.
Oh no, not to worry. Thanks for your concern, though. It is very kind.
Perhaps undeserved, though. The real realization is that even knowing this, I will not change my actions, i.e. it isn't ignorance. The practice of what I do is that I value an arbitrary human life pretty lowly.
I think if you gave me a live feed of a child dying, I would certainly pay the $5k to stop the child from dying, or the $10k, or the whatever. But even the slightest distance and I wouldn't send $20 over. Odd. I can even visualize the child and it feels a bit bad but I'll still take the super burrito.
Personally I have a hard time giving an arbitrary sum, but setting up an automated donation each month that I can forget worked for me, it helps remove the decision aspect.
I suppose buying life-saving gear and similar spending is only a medium-sized fraction of the foundation's spending. I wonder what the accurate expected value of lives saved per dollar donated is.
You can see the spreadsheet model GiveWell used to come up with their "dollars per life saved" number here. See the "Bednets" tab for the AMF calculation. With regard to how the foundation spends its money, the only input considered is "dollars donated per anti-malarial bed net distributed" (about five bucks, depending on country), and work proceeds from there to establish the value of one bed net distribution.
https://www.givewell.org/charities/amf
You might also want to read this meta-discussion by GiveWell on such analyses:
https://blog.givewell.org/2011/08/18/why-we-cant-take-expect...