“As far as how to optimally distribute money to charity, that is very much an unsolved problem, but I think it’s one that we can mostly worry about when we get that far.”
I like the rest of your proposal, but I seriously think we need to look more carefully at this part. Once a billion dollars is already on the line, it’s worthwhile for large charities that won’t do very much good to spend $100M on marketing for a 12% chance at getting it, which does no one any good at all (except the marketing companies). If we make the decision beforehand- even if it is completely arbitrary (eg., we take all the charities recommended by GiveWell and put them on a giant roulette wheel)- then charities won’t spend large amounts of money competing amongst themselves for the money, which would defeat the original purpose.
Once a billion dollars is already on the line, it’s worthwhile for large charities that won’t do very much good to spend $100M on marketing for a 12% chance at getting it, which does no one any good at all (except the marketing companies).
Steve Rayhawk once showed me an article claiming that (nearly) all of the value of the land on the American frontier literally vanished because of considerations like these. It was one of the most horrifying things I’d ever seen.
In my opinion, the easiest way to handle this is to pick a handful of charities (with the input and approval of Buckmaster) and have the Facebook group vote on which they like.
It gives them the impression of control but since you picked the options on the poll you can guide it to somewhere that will do something useful with it.
I have not been convinced that this scenario or various permutations is not what we should go for. I think the chance of failure goes up dramatically if we do not allow for some sort of ongoing, annual vote for the charity distribution. We will have the bully pulpit of a Facebook page and mailing list to teach our fellow activists about rational giving, but the choice should be made by the people. Craig and Jim like to think they are not dictators and picking the charities and the division of money between them will probably go against their sense of fairness.
One charity I know Craig supports is Kiva. Is Kiva really that bad? I’ve heard some backlash against them recently. Craig also seems to be involved with various anti-corruption causes; that is something I would be ok getting behind that beats “The Society for Rare Diseases in Photogenic Puppies”.
Craig and Jim like to think they are not dictators and picking the charities and the division of money between them will probably go against their sense of fairness.
True, but technically this is their money we are giving away.
I personally don’t know much about charities. I just think it is an easier sell if we pick names that people recognize and let them choose from the ones we consider the best. By “we” I mean someone around here who knows more about charities.
The easiest way to handle this is to pick the names of potential charities now with Craig and Jim’s approval and put all of them on the website with details about how we will choose between them. The actual vote will happen after the Facebook group succeeds.
This does multiple things: (a) we get to use more names for charities, (b) people who have an aversion to specific charities will have an incentive to vote but will not be discouraged from joining, (c) locks in potential charities so we can avoid golddiggers sniffing around after success.
So, for example, we say:
We have been impressed with the work of charities X, Y, Z and are planning on donating the proceeds from this venture to them. After we succeed we will have a vote amongst the Facebook group to determine exactly how to divvy up the massive amount of money you raised.
Good point, Tom. Of course, even better than picking something completely arbitrary would be to steer the money towards the causes that matter most. Off the top of my head I would say SIAI, the Methuselah Foundation, and the Lifeboat Foundation. Unfortunately I don’t think any of these would be able to attract 50,000 supporters, but maybe there’s some way we could package them with other charities.
Sadly, I don’t think existential risk reduction is sufficiently sympathetic to the general population (and we do need them on board for this to work). And if you have a large basket with stuff like the Methuselah foundation in it, you’re likely to have people wondering why they can’t put in “The Society for Rare Diseases in Photogenic Puppies”.
Ideally, you’d pick something simple and widely acceptable. Obviously, it would be difficult to find a single charity that could productively use a billion extra dollars per year. But the basket should be as simple and uncontroversial (and obviously, productive) as possible).
Edit: Thinking about it, using a trusted intermediary might make the most sense. Using a grant-making agency avoids the appearance that we’re funneling the money to our pet causes, it reduces the marketing/lobbying incentives (though it doesn’t eliminate them) and it makes the money relatively productive (if we choose a good agency). Givewell may be a poor choice, due to the Metafilter flap, but we could specify, say the MIT Poverty Action Lab or something.
Obviously, we’d need the organizations cooperation, or at least permission.
I would like to see this bundled with a Rational Charity meme. Let’s be frank here: if this ends up going to the Society for Rare Diseases in Photogenic Puppies, it wasn’t worth LW’s time. If we can manage to get some money to things that actually matter, it was.
Trying to get something worthwhile done, as opposed to “making a billion dollars go to charity”, might make the whole project fail because of that added extra inconvenience. So what?
If you wanted to boil it down to a meme, it would be “Do something effective for a change”. Supposing you actually can generate a billion dollars, that’s enough for ten million dollars for one hundred charities. “Ten million dollars apiece for one hundred unusual and effective charities.” Like that.
In the past, you’ve pointed out that it can never be more efficient to split a small donation between two charities than to give all to the best bet, even if you are uncertain which is best. So I take it the advantage of lots of charities here is a political one, that we can include some sops to fuzzy-purchasing, lots of GiveWell-ish charities whose efficiency we can calculate, and perhaps one or two x-risk charities which we consider to be very efficient but which most people aren’t sold on?
True :-) But is it really so much that in order not to reach diminishing returns on an individual charity, it has to be split 100 ways? Even splitting it five ways would seem to be enough to offset that effect.
As long as some amount of it goes to worthwhile charities, I think the whole thing would be worthwhile. I think we’d be hard pressed to lose control of the meme such that of a billion dollars at least 10 million didn’t go to charities that we would want to support.
I think it ought to be something unimaginative but reliable, like clean water or vaccines to third world countries. I can’t find it at the moment but there’s a highly reputable charity that provides clean drinking water to African communities. IIRC they estimated that every $400 or so saved the life of a child. A billion dollars into such a charity—saving 2.5 million children—isn’t a difficult PR sell.
The problem is not finding an effective, productive, and reputable charity. There are plenty out there (even if a majority are not). It’s finding a charity than can effectively and productively use an extra billion dollars. Many charities don’t have the oversight and planning infrastructure to use a windfall of that size.
Most of that is given to churches, hospitals, rich-country education, etc. Much, much less is given to overseas public health aid, and less of that to efficient programs.
Mimicking the Gates Foundation grants to GAVI could absorb a lot, but would risk missing a lot of the potential to use this to promote more efficient giving.
I was thinking yesterday about what I’d do if I won the lottery. Then I recalled that Robin Hanson gave the best version of the “What I’d do with a million” story I’ve yet heard, so I figured I’d give it to him. I don’t know what he’d do with a billion, but I’d still point to his plan as one of the top ways to increase efficiency through spending money.
It seems to me that to do this effectively, Less Wrong will need to make a lot of good decisions. Upvoting and downvoting comments works for deciding what quotes should appear first in a rationality quote thread, but to do something as important as choosing which charities to promote or choosing the name of a Facebook group, I think we could use something higher-caliber.
Range voting looks pretty good but maybe someone who has studied voting systems can suggest something better. I’m thinking maybe a web app that would make it easy to create polls, add options, and vote on them, maybe with captchas associated with all those actions to discourage any lurking 4chan types. It’s 2 AM in my time zone, and I’m below the journeyman level in web development, so perhaps someone else would like to cook this up. If not I’ll probably work on it tomorrow.
“As far as how to optimally distribute money to charity, that is very much an unsolved problem, but I think it’s one that we can mostly worry about when we get that far.”
I like the rest of your proposal, but I seriously think we need to look more carefully at this part. Once a billion dollars is already on the line, it’s worthwhile for large charities that won’t do very much good to spend $100M on marketing for a 12% chance at getting it, which does no one any good at all (except the marketing companies). If we make the decision beforehand- even if it is completely arbitrary (eg., we take all the charities recommended by GiveWell and put them on a giant roulette wheel)- then charities won’t spend large amounts of money competing amongst themselves for the money, which would defeat the original purpose.
Steve Rayhawk once showed me an article claiming that (nearly) all of the value of the land on the American frontier literally vanished because of considerations like these. It was one of the most horrifying things I’d ever seen.
I would love to see said article.
Scroll about halfway down to “The Most Expensive Mistake”
http://www.daviddfriedman.com/Laws_Order_draft/laws_order_ch_10.htm
In my opinion, the easiest way to handle this is to pick a handful of charities (with the input and approval of Buckmaster) and have the Facebook group vote on which they like.
It gives them the impression of control but since you picked the options on the poll you can guide it to somewhere that will do something useful with it.
I have not been convinced that this scenario or various permutations is not what we should go for. I think the chance of failure goes up dramatically if we do not allow for some sort of ongoing, annual vote for the charity distribution. We will have the bully pulpit of a Facebook page and mailing list to teach our fellow activists about rational giving, but the choice should be made by the people. Craig and Jim like to think they are not dictators and picking the charities and the division of money between them will probably go against their sense of fairness.
One charity I know Craig supports is Kiva. Is Kiva really that bad? I’ve heard some backlash against them recently. Craig also seems to be involved with various anti-corruption causes; that is something I would be ok getting behind that beats “The Society for Rare Diseases in Photogenic Puppies”.
True, but technically this is their money we are giving away.
I personally don’t know much about charities. I just think it is an easier sell if we pick names that people recognize and let them choose from the ones we consider the best. By “we” I mean someone around here who knows more about charities.
The easiest way to handle this is to pick the names of potential charities now with Craig and Jim’s approval and put all of them on the website with details about how we will choose between them. The actual vote will happen after the Facebook group succeeds.
This does multiple things: (a) we get to use more names for charities, (b) people who have an aversion to specific charities will have an incentive to vote but will not be discouraged from joining, (c) locks in potential charities so we can avoid golddiggers sniffing around after success.
So, for example, we say:
http://www.philanthropyaction.com/nc/a_mostly_comprehensive_guide_to_the_kiva_and_donor_illusion_debate
Good point, Tom. Of course, even better than picking something completely arbitrary would be to steer the money towards the causes that matter most. Off the top of my head I would say SIAI, the Methuselah Foundation, and the Lifeboat Foundation. Unfortunately I don’t think any of these would be able to attract 50,000 supporters, but maybe there’s some way we could package them with other charities.
Sadly, I don’t think existential risk reduction is sufficiently sympathetic to the general population (and we do need them on board for this to work). And if you have a large basket with stuff like the Methuselah foundation in it, you’re likely to have people wondering why they can’t put in “The Society for Rare Diseases in Photogenic Puppies”.
Ideally, you’d pick something simple and widely acceptable. Obviously, it would be difficult to find a single charity that could productively use a billion extra dollars per year. But the basket should be as simple and uncontroversial (and obviously, productive) as possible).
Edit: Thinking about it, using a trusted intermediary might make the most sense. Using a grant-making agency avoids the appearance that we’re funneling the money to our pet causes, it reduces the marketing/lobbying incentives (though it doesn’t eliminate them) and it makes the money relatively productive (if we choose a good agency). Givewell may be a poor choice, due to the Metafilter flap, but we could specify, say the MIT Poverty Action Lab or something.
Obviously, we’d need the organizations cooperation, or at least permission.
I would like to see this bundled with a Rational Charity meme. Let’s be frank here: if this ends up going to the Society for Rare Diseases in Photogenic Puppies, it wasn’t worth LW’s time. If we can manage to get some money to things that actually matter, it was.
Trying to get something worthwhile done, as opposed to “making a billion dollars go to charity”, might make the whole project fail because of that added extra inconvenience. So what?
If you wanted to boil it down to a meme, it would be “Do something effective for a change”. Supposing you actually can generate a billion dollars, that’s enough for ten million dollars for one hundred charities. “Ten million dollars apiece for one hundred unusual and effective charities.” Like that.
In the past, you’ve pointed out that it can never be more efficient to split a small donation between two charities than to give all to the best bet, even if you are uncertain which is best. So I take it the advantage of lots of charities here is a political one, that we can include some sops to fuzzy-purchasing, lots of GiveWell-ish charities whose efficiency we can calculate, and perhaps one or two x-risk charities which we consider to be very efficient but which most people aren’t sold on?
We’re not talking about a small donation.
True :-) But is it really so much that in order not to reach diminishing returns on an individual charity, it has to be split 100 ways? Even splitting it five ways would seem to be enough to offset that effect.
Unless one of the charities is SingInst.
As long as some amount of it goes to worthwhile charities, I think the whole thing would be worthwhile. I think we’d be hard pressed to lose control of the meme such that of a billion dollars at least 10 million didn’t go to charities that we would want to support.
I think it ought to be something unimaginative but reliable, like clean water or vaccines to third world countries. I can’t find it at the moment but there’s a highly reputable charity that provides clean drinking water to African communities. IIRC they estimated that every $400 or so saved the life of a child. A billion dollars into such a charity—saving 2.5 million children—isn’t a difficult PR sell.
The problem is not finding an effective, productive, and reputable charity. There are plenty out there (even if a majority are not). It’s finding a charity than can effectively and productively use an extra billion dollars. Many charities don’t have the oversight and planning infrastructure to use a windfall of that size.
There is an obvious solution to this: fund multiple charities.
Philanthropy by Americans alone is about $300 billion per year. The guesstimated annual cashflow here is less than one-thousandth of that.
Most of that is given to churches, hospitals, rich-country education, etc. Much, much less is given to overseas public health aid, and less of that to efficient programs.
Mimicking the Gates Foundation grants to GAVI could absorb a lot, but would risk missing a lot of the potential to use this to promote more efficient giving.
I was thinking yesterday about what I’d do if I won the lottery. Then I recalled that Robin Hanson gave the best version of the “What I’d do with a million” story I’ve yet heard, so I figured I’d give it to him. I don’t know what he’d do with a billion, but I’d still point to his plan as one of the top ways to increase efficiency through spending money.
It seems to me that to do this effectively, Less Wrong will need to make a lot of good decisions. Upvoting and downvoting comments works for deciding what quotes should appear first in a rationality quote thread, but to do something as important as choosing which charities to promote or choosing the name of a Facebook group, I think we could use something higher-caliber.
Range voting looks pretty good but maybe someone who has studied voting systems can suggest something better. I’m thinking maybe a web app that would make it easy to create polls, add options, and vote on them, maybe with captchas associated with all those actions to discourage any lurking 4chan types. It’s 2 AM in my time zone, and I’m below the journeyman level in web development, so perhaps someone else would like to cook this up. If not I’ll probably work on it tomorrow.
Edit: Looks like something similar already exists, see http://vote.superduperapps.com/