It still seems possible to save QALYs for a few hundred dollars in the developing world, whereas the UK’s NHS is willing to fund most things that save a QALY for under £20,000, and some that are over £30,000, which is again a factor of 100 difference.
I like these examples but they do have some limitations.
I’m still searching for some better examples that are empirically robust as well as intuitively powerful.
(I’m looking for the strongest references to give to my claim here that “there is a strong case that most donations go to charities that improve well-being far less per-dollar than others.” (of course I’m willing to admit there’s some possibility that we don’t have strong evidence for this)
1) Differences in income: This will not be terribly convincing to anyone who doesn’t already accept the idea of vastly diminishing marginal utility, and there is the standard (inadequate but hard to easily rebut) objection that “things are much cheaper in developing countries”.
2) The cost to save a life: Yes, rich country governments factor this into their calculations, but is this indeed the calculation that is relevant when considering “typical charities operating in rich countries?” It also does not identify a particular intervention that is “much less efficient”.
3) Cost per QALY/ UK NHS: Similar limitations as in case 2.
What is the strongest statistic or comparison for making this point? Perhaps Sanjay Joshi of SoGive has some suggestions?
Hi Peter,
Some examples that might be useful:
1) Differences in income
A US college graduate earns about 100x more than GiveDirectly recipients, suggesting money can go far further with GiveDirectly. (100x further if utility ~log-income.) https://80000hours.org/career-guide/anyone-make-a-difference/
2) The cost to save a life
GiveWell now says $7500 for a death prevented by malaria nets (plus many other benefits) Rich country governments, however, are often willing to pay over $1m to save a life of one of their citizens, a factor of 130+ difference. https://80000hours.org/career-guide/world-problems/#global-health-a-problem-where-you-could-really-make-progress
3) Cost per QALY
It still seems possible to save QALYs for a few hundred dollars in the developing world, whereas the UK’s NHS is willing to fund most things that save a QALY for under £20,000, and some that are over £30,000, which is again a factor of 100 difference.
So I still think a factor of 100x difference is defensible, though if you also take into account Brian’s point below, then it might be reduced to, say, a factor of 30, though that’s basically just a guess, and it could go the other way too. More on this: http://reflectivedisequilibrium.blogspot.com/2014/01/what-portion-of-boost-to-global-gdp.html
I like these examples but they do have some limitations.
I’m still searching for some better examples that are empirically robust as well as intuitively powerful.
(I’m looking for the strongest references to give to my claim here that “there is a strong case that most donations go to charities that improve well-being far less per-dollar than others.” (of course I’m willing to admit there’s some possibility that we don’t have strong evidence for this)
1) Differences in income: This will not be terribly convincing to anyone who doesn’t already accept the idea of vastly diminishing marginal utility, and there is the standard (inadequate but hard to easily rebut) objection that “things are much cheaper in developing countries”.
2) The cost to save a life: Yes, rich country governments factor this into their calculations, but is this indeed the calculation that is relevant when considering “typical charities operating in rich countries?” It also does not identify a particular intervention that is “much less efficient”.
3) Cost per QALY/ UK NHS: Similar limitations as in case 2.
What is the strongest statistic or comparison for making this point? Perhaps Sanjay Joshi of SoGive has some suggestions?
Perhaps making a comparison based on the tables near the end of Jamison, D. T. et al (2006). Disease control priorities in developing countries? 2006 was a long time ago, however.