Why haven’t the good people at GiveWell written more about anti-aging research?
According to GiveWell, the AMF can save a life for $3.4e3. Let’s say it’s a young life with 5e1 years to live. A year is 3.1e7 seconds, so saving a life gives humanity 1.5e9 seconds, or about 5e5 sec/$.
Suppose you could invest $1e6 in medical research to buy a 50-second increase in global life expectancy. Approximating global population as 1e10, this buys humanity 5e11 seconds, or about the same value of 5e5 sec/$.
Buying a 50-second increase in life expectancy for a megabuck seems very doable. In practice, any particular medical innovation wouldn’t give 50 seconds to everyone, but instead would give a larger chunk of time (say, a week) to a smaller number of people suffering from a specific condition. But the math could work out the same.
Of course, it could turn out that the cost of extending humanity’s aggregate lifespan with medical research is much more than $5e5/sec. But it could also turn out to be much cheaper than that.
ETA: GiveWell has in fact done a lot of research on this theme, thanks to ChristianKl for pointing this out below.
For AMF it’s a lot easier to estimate the effect than it is for anti-aging research.
GiveWell purposefully started with a focus on interventions for which the can study the effect.
GiveWell writes:
Medical research : As of November 2011, we are just beginning to consider the cause of medical research. Conceptually, we find this cause promising because it is possible that a relatively small amount spent on research and development could result in new disease-fighting technology that could be used to save and improve many lives throughout the world. However, we do not yet have a good sense of whether this cause has a strong track record of turning charitable dollars into lives saved and improved.
“What are the best opportunities for funders aiming to contribute to progress in life sciences (i.e., biology and medicine)?” This post lays out what we’ve done to date and how we plan to move forward.
“What are the best opportunities for funders aiming to contribute to progress in life sciences (i.e., biology and medicine)?” This post lays out what we’ve done to date and how we plan to move forward.
GiveWell Labs managed get
Steve Goodman and John Ioannidis matchmaked with the Laura and John Arnold Foundation at the tune of $6 million.
Meta-Research doesn’t sound as sexy as anti-aging research but if we want to have good anti-aging research we need a good basis in biology as a whole.
Anti-aging research is a catch-phrase and it makes sense that it’s decently funded but alone it won’t work. Biology as a whole needs to progress and chasing after shiny anti-aging targets might not always be the most effective use of money.
Do you have a reason why you think it makes more sense to speak about anti-aging research than it makes sense to speak about life-science research?
Buying a 50-second increase in life expectancy for a megabuck seems very doable.
Please do a Fermi estimation of how you arrive at that conclusion.
Ooh, I know! So, Holden is aware of SENS. However, by default, GiveWell doesn’t publish any info on charities it looks at and decides not to recommend, if they don’t ask GiveWell to. This is to encourage other charities to go through GiveWell’s recommendation process—it keeps GiveWell from lowering a charity’s reputation by evaluating them.
Anyways, GiveWell did some sort of surface-level look at SENS a while back, and didn’t recommend them. I think the only way to get more info about this would be to email Aubrey about his interaction with GiveWell.
When doing the calculations be sure to QA your LYs. Spending an extra week lying doped up and in pain in a hospital bed may not be worth all that much. Also with medical research, you often wind up with a patented drug which then costs $1e5 per patient treated at least for the first decade or two of its use at least as used in the USA and other non-single payer countries. Or it requires $1e5 of medical professional intervention per patient to implement. My priors are that the low-hanging fruit is not in turning 90 year olds into 91 year olds, and won’t be for many decades.
I think their argument was that they don’t support Pascal’s Mugging and they don’t see any proof of medical research within reach that could end aging with a significant probability.
EDIT: …and I should have read the comment in more detail. You are talking about stuff such as donating to curing diseases. I think they just didn’t assign analysts to this yet. I guess it’s hard to measure scientific progress.
Why haven’t the good people at GiveWell written more about anti-aging research?
According to GiveWell, the AMF can save a life for $3.4e3. Let’s say it’s a young life with 5e1 years to live. A year is 3.1e7 seconds, so saving a life gives humanity 1.5e9 seconds, or about 5e5 sec/$.
Suppose you could invest $1e6 in medical research to buy a 50-second increase in global life expectancy. Approximating global population as 1e10, this buys humanity 5e11 seconds, or about the same value of 5e5 sec/$.
Buying a 50-second increase in life expectancy for a megabuck seems very doable. In practice, any particular medical innovation wouldn’t give 50 seconds to everyone, but instead would give a larger chunk of time (say, a week) to a smaller number of people suffering from a specific condition. But the math could work out the same.
Of course, it could turn out that the cost of extending humanity’s aggregate lifespan with medical research is much more than $5e5/sec. But it could also turn out to be much cheaper than that.
ETA: GiveWell has in fact done a lot of research on this theme, thanks to ChristianKl for pointing this out below.
For AMF it’s a lot easier to estimate the effect than it is for anti-aging research. GiveWell purposefully started with a focus on interventions for which the can study the effect.
GiveWell writes:
You find a bit of data gathering under http://www.givewell.org/node/1339
More recently GiveWell Labs which then was renamed into the Open Philanthropy project will put more emphasis in that direction.
Articles that were written are:
http://blog.givewell.org/2013/12/26/scientific-research-funding/
http://blog.givewell.org/2014/01/07/exploring-life-sciences-funding/
http://blog.givewell.org/2014/01/15/returns-to-life-sciences-funding/
GiveWell Labs managed get Steve Goodman and John Ioannidis matchmaked with the Laura and John Arnold Foundation at the tune of $6 million.
Meta-Research doesn’t sound as sexy as anti-aging research but if we want to have good anti-aging research we need a good basis in biology as a whole.
Anti-aging research is a catch-phrase and it makes sense that it’s decently funded but alone it won’t work. Biology as a whole needs to progress and chasing after shiny anti-aging targets might not always be the most effective use of money. Do you have a reason why you think it makes more sense to speak about anti-aging research than it makes sense to speak about life-science research?
Please do a Fermi estimation of how you arrive at that conclusion.
Ooh, I know! So, Holden is aware of SENS. However, by default, GiveWell doesn’t publish any info on charities it looks at and decides not to recommend, if they don’t ask GiveWell to. This is to encourage other charities to go through GiveWell’s recommendation process—it keeps GiveWell from lowering a charity’s reputation by evaluating them.
Anyways, GiveWell did some sort of surface-level look at SENS a while back, and didn’t recommend them. I think the only way to get more info about this would be to email Aubrey about his interaction with GiveWell.
When doing the calculations be sure to QA your LYs. Spending an extra week lying doped up and in pain in a hospital bed may not be worth all that much. Also with medical research, you often wind up with a patented drug which then costs $1e5 per patient treated at least for the first decade or two of its use at least as used in the USA and other non-single payer countries. Or it requires $1e5 of medical professional intervention per patient to implement. My priors are that the low-hanging fruit is not in turning 90 year olds into 91 year olds, and won’t be for many decades.
I think their argument was that they don’t support Pascal’s Mugging and they don’t see any proof of medical research within reach that could end aging with a significant probability.
EDIT: …and I should have read the comment in more detail. You are talking about stuff such as donating to curing diseases. I think they just didn’t assign analysts to this yet. I guess it’s hard to measure scientific progress.