I could imagine that happening in some situations where after a lot of careful thought you decide to defer to another grantmaker, but if you know in advance that you’d like to give your money to a grantmaker, shouldn’t you just do that?
Yeah you probably should—unless perhaps you think there are scale effects to giving which makes you want to punt on being able to give far more.
Worth noting of course that Patrick didn’t know he was going to give to a capital allocator when he entered the lottery though, and of course still doesn’t. Ideally all donor lottery winners would examine the LTFF very carefully and honestly consider whether they think they can do better than LTFF. People may be able to beat LTFF, but if someone isn’t giving to LTFF I would expect clear justification as to why they think they can beat it.
I disagree. One of the original rationales for the lottery if I recall correctly was to increase the diversity* of funding sources and increase the number of grantmakers. I think if the LTFF is particularly funding constrained, there’s a good chance the Open Philanthropy Project or a similar organisation will donate to them. I value increased diversity and number of grantmakers enough that I think it’s worth trying to beat LTFF’s grantmaking even if you might fail.
*By diversity, I don’t mean gender or ethnicity, I just mean having more than one grantmaker doing the same thing, ideally with different knowledge, experience and connections.
I’m not sure I understand how the lottery increases the diversity of funding sources / increases the number of grantmakers if one or a small number of people end up winning the lottery. Wouldn’t it actually reduce diversity / number of grantmakers? I might be missing something quite obvious here...
Reading this it seems the justification for lotteries is that it not only saves research time for the EA community as a whole, but also improves the allocation of the money in expectation. Basically if you don’t win you don’t have to bother doing any research (so this time is saved for lots of people), and if you do win you at least have the incentive to do lots of research because you’re giving away quite a lot of money (so the money should be given away with a great deal of careful thought behind it).
Of course if everyone in the EA community just gives to an EA Fund and knows that they would do so if they won the lottery, that would render both of the benefits of the lottery redundant. This shouldn’t be the case however as A) not everyone gives to EA Funds—some people really research where they give, and B) people playing donor lotteries shouldn’t be certain of where they would give the money if they won—the idea is that they would have to research. I see no reason why this research shouldn’t lead to giving to an EA Fund.
I think perhaps we agree then—if after significant research, you realize you can’t beat an EA Fund, that seems like a reasonable fallback, but that should not be plan A.
Re: increasing grantmakers, I meant increasing the number of grantmakers who have spent significant time thinking about where to donate significant capital—obviously having hundreds of people donating $1k each would have more diversity but in practice I think most $1k donors defer their decision-making to someone else, like an EA Fund or GiveWell.
I think perhaps we agree then—if after significant research, you realize you can’t beat an EA Fund, that seems like a reasonable fallback, but that should not be plan A.
Yeah that sounds about right to me.
I meant increasing the number of grantmakers who have spent significant time thinking about where to donate significant capital
I still don’t understand this. The lottery means one / a small number of grantmakers get all the money to allocate. People who don’t win don’t need to think about where to donate. So really it seems to me that the lottery reduces the number of grantmakers and indeed the number of who spend time thinking about where to donate.
I still don’t understand this. The lottery means one / a small number of grantmakers get all the money to allocate. People who don’t win don’t need to think about where to donate. So really it seems to me that the lottery reduces the number of grantmakers and indeed the number of who spend time thinking about where to donate.
The model is this:
A bunch of people each have $5,000 to donate.
Many put in a bit of effort—they spend a bit of time on the GiveWell website, read some stuff by MIRI, and chat to a couple of friends. But this isn’t enough to catch them up on the state of the art, let alone make some novel contribution to the grant application discrimination project.
Others can’t find the time to do even this much research.
So overall very little grant evaluation has really been done, and what has been done is highly duplicative. Given they all fail to pass the bar of ‘as good as the EA funds’, this work was essentially wasted.
But if they instead did a lottery:
One person gets $500,000 to donate.
He now puts in a lot of effort—reading a huge amount of literature, and doing calls with the leaders of multiple organizations. Perhaps he also discusses his approaches with several other EAs for advice.
By the end he has a novel understanding of some aspect of the charitable funding landscape, which exceeds that of the EA fund grantmakers.
The overall amount of time spent is actually less than before, but the depth is far greater, and with dramatically less redundancy.
So by using the lottery we have both saved time and increased the amount of effective evaluation work being done.
I think deciding between capital allocators is a great use of the donor lottery, even as a Plan A. You might say something like: “I would probably give to the Long-Term Future Fund, but I’m not totally sure whether they’re better than the EA Infrastructure Fund or Longview or something I might come up with myself. So I’ll participate in the donor lottery so if I win, I can take more time to read their reports and see which of them seems best.” I think this would be a great decision.
I’d be pretty unhappy if such a donor then felt forced to instead do their own grantmaking despite not having a comparative advantage for doing so (possibly underperforming Open Phil’s last dollar), or didn’t participate in the donor lottery in the first place. I think the above use case is one of the most central one that I hope to address.
I tentatively agree that further diversification of funding sources might be good, but I don’t think the donor lottery is the right tool for that.
I could imagine that happening in some situations where after a lot of careful thought you decide to defer to another grantmaker, but if you know in advance that you’d like to give your money to a grantmaker, shouldn’t you just do that?
Yeah you probably should—unless perhaps you think there are scale effects to giving which makes you want to punt on being able to give far more.
Worth noting of course that Patrick didn’t know he was going to give to a capital allocator when he entered the lottery though, and of course still doesn’t. Ideally all donor lottery winners would examine the LTFF very carefully and honestly consider whether they think they can do better than LTFF. People may be able to beat LTFF, but if someone isn’t giving to LTFF I would expect clear justification as to why they think they can beat it.
I disagree. One of the original rationales for the lottery if I recall correctly was to increase the diversity* of funding sources and increase the number of grantmakers. I think if the LTFF is particularly funding constrained, there’s a good chance the Open Philanthropy Project or a similar organisation will donate to them. I value increased diversity and number of grantmakers enough that I think it’s worth trying to beat LTFF’s grantmaking even if you might fail.
*By diversity, I don’t mean gender or ethnicity, I just mean having more than one grantmaker doing the same thing, ideally with different knowledge, experience and connections.
I’m not sure I understand how the lottery increases the diversity of funding sources / increases the number of grantmakers if one or a small number of people end up winning the lottery. Wouldn’t it actually reduce diversity / number of grantmakers? I might be missing something quite obvious here...
Reading this it seems the justification for lotteries is that it not only saves research time for the EA community as a whole, but also improves the allocation of the money in expectation. Basically if you don’t win you don’t have to bother doing any research (so this time is saved for lots of people), and if you do win you at least have the incentive to do lots of research because you’re giving away quite a lot of money (so the money should be given away with a great deal of careful thought behind it).
Of course if everyone in the EA community just gives to an EA Fund and knows that they would do so if they won the lottery, that would render both of the benefits of the lottery redundant. This shouldn’t be the case however as A) not everyone gives to EA Funds—some people really research where they give, and B) people playing donor lotteries shouldn’t be certain of where they would give the money if they won—the idea is that they would have to research. I see no reason why this research shouldn’t lead to giving to an EA Fund.
I think perhaps we agree then—if after significant research, you realize you can’t beat an EA Fund, that seems like a reasonable fallback, but that should not be plan A.
Re: increasing grantmakers, I meant increasing the number of grantmakers who have spent significant time thinking about where to donate significant capital—obviously having hundreds of people donating $1k each would have more diversity but in practice I think most $1k donors defer their decision-making to someone else, like an EA Fund or GiveWell.
Yeah that sounds about right to me.
I still don’t understand this. The lottery means one / a small number of grantmakers get all the money to allocate. People who don’t win don’t need to think about where to donate. So really it seems to me that the lottery reduces the number of grantmakers and indeed the number of who spend time thinking about where to donate.
The model is this:
A bunch of people each have $5,000 to donate.
Many put in a bit of effort—they spend a bit of time on the GiveWell website, read some stuff by MIRI, and chat to a couple of friends. But this isn’t enough to catch them up on the state of the art, let alone make some novel contribution to the grant application discrimination project.
Others can’t find the time to do even this much research.
So overall very little grant evaluation has really been done, and what has been done is highly duplicative. Given they all fail to pass the bar of ‘as good as the EA funds’, this work was essentially wasted.
But if they instead did a lottery:
One person gets $500,000 to donate.
He now puts in a lot of effort—reading a huge amount of literature, and doing calls with the leaders of multiple organizations. Perhaps he also discusses his approaches with several other EAs for advice.
By the end he has a novel understanding of some aspect of the charitable funding landscape, which exceeds that of the EA fund grantmakers.
The overall amount of time spent is actually less than before, but the depth is far greater, and with dramatically less redundancy.
So by using the lottery we have both saved time and increased the amount of effective evaluation work being done.
Thanks, I understand all that. I was confused when Khorton said:
I wouldn’t say the lottery increases the number of grantmakers who have spent significant time thinking, I think it in fact reduces it.
I agree with you when you say however:
I think deciding between capital allocators is a great use of the donor lottery, even as a Plan A. You might say something like: “I would probably give to the Long-Term Future Fund, but I’m not totally sure whether they’re better than the EA Infrastructure Fund or Longview or something I might come up with myself. So I’ll participate in the donor lottery so if I win, I can take more time to read their reports and see which of them seems best.” I think this would be a great decision.
I’d be pretty unhappy if such a donor then felt forced to instead do their own grantmaking despite not having a comparative advantage for doing so (possibly underperforming Open Phil’s last dollar), or didn’t participate in the donor lottery in the first place. I think the above use case is one of the most central one that I hope to address.
I tentatively agree that further diversification of funding sources might be good, but I don’t think the donor lottery is the right tool for that.