Once upon a time, I found myself with a bunch of unconditional $25 charity gift cards from an every.org promotion. This seemed like a great opportunity to encourage the people in my life to pick charities to donate to, without the awkwardness of talking directly about how they should spend their own money. So I sent four gift card links and an explanation to a group chat with my four closest friends from college.
The first thing that happened was that one friend expressed enthusiasm, claimed a gift card, and donated it to the Florence Project, an organization that gives legal aid to detained migrants in Arizona. The second thing that happened was that the other three friends said nothing, and never claimed or used the gift cards.
I felt disappointed with this response rate. I mean, why wouldn’t someone want a free $25 for charity?? I know on an intellectual level that I’m much more excited about picking charities to donate to than most people are, but do I really have to internalize that knowledge in order to make accurate predictions? Darn.
I didn’t want to pressure anyone, so I didn’t say any more about it. As time passed and the gift cards drew close to their expiration date, I figured I should just use them myself. But here I ran into an issue. If I just directed this unclaimed money to the organizations that I would have chosen in the first place, wasn’t I kind of retroactively giving myself reason to prefer that my friends not participate? I didn’t want that! At the time I had genuinely wanted my friends to participate, and I didn’t want to ruin that after the fact!
Of course, I also didn’t want to waste $75 of donations for no reason. Or for dubious decision-theoretic reasons that definitely had not been relevant to the actual behavior of any of the actual people involved.
And so, this has been the convoluted story how I came to donate a particular $75 to GiveDirectly. Usually, I don’t donate to GiveDirectly, because I’m on board with GiveWell’s view that donations to AMF and other top charities are something like ten times as cost-effective as unconditional cash transfers. (See e.g. this funding report.) But I’m a big fan of cash benchmarking—the idea that all global development interventions should be compared to the baseline option of just giving the beneficiaries unconditional cash transfers instead. Similarly, it seemed to me that if money was earmarked for charity, and no one stepped in to make a case for anything else, cash was a sensible default.
I think this approach is potentially useful in a much more common situation: the matching campaign. Matching campaigns are popular, but misleading (see e.g. discussion here, here, and here). The misleadingness could be fixed if the matching donor agreed to set fire to any money that wasn’t used to match other donations, but they typically won’t agree to that (and it wouldn’t be a great look). The idea here is, instead of fire, the matching donor picks a “floor” donation option that they’re okay with their money going to, but not maximally excited about. For best results, the audience of potential donors should largely agree with this assessment, and also the floor option should have some kind of ontological basicness to it, such that it feels reasonable and not insulting to designate it as the “floor”. Obviously I’m describing cash transfers here, but I think other interventions with these properties would work for the same reasons; direct air capture comes to mind.
I’d be interested to hear if something like this has been tried.
Interesting suggestion. I’m not familiar with anyone doing a donation match like this.
It seems like having a default charity for matching money to go to could be counterproductive to the matcher’s goals. E.g. Every.org wanted to get more people to use their platform to donate. But I think many people don’t really find it more valuable for money to get directed to one charity over another. EAs are different in that regard. While we’re certainly not unique in caring which charities money goes to, I think many people might think “Why should I donate when the money is already going to go to charity?” and decide not to participate.
While generally I wouldn’t advise people to do donation matches, would it be good for organizations already running them to make cash transfers the default use of the money if matching donors don’t direct it elsewhere? Maybe. One benefit might be that it just gets people to think more about the value of directing money to one organization versus another, instead of merely thinking that they’re raising more money for a charity of their choice.
Cash by Default
Once upon a time, I found myself with a bunch of unconditional $25 charity gift cards from an every.org promotion. This seemed like a great opportunity to encourage the people in my life to pick charities to donate to, without the awkwardness of talking directly about how they should spend their own money. So I sent four gift card links and an explanation to a group chat with my four closest friends from college.
The first thing that happened was that one friend expressed enthusiasm, claimed a gift card, and donated it to the Florence Project, an organization that gives legal aid to detained migrants in Arizona. The second thing that happened was that the other three friends said nothing, and never claimed or used the gift cards.
I felt disappointed with this response rate. I mean, why wouldn’t someone want a free $25 for charity?? I know on an intellectual level that I’m much more excited about picking charities to donate to than most people are, but do I really have to internalize that knowledge in order to make accurate predictions? Darn.
I didn’t want to pressure anyone, so I didn’t say any more about it. As time passed and the gift cards drew close to their expiration date, I figured I should just use them myself. But here I ran into an issue. If I just directed this unclaimed money to the organizations that I would have chosen in the first place, wasn’t I kind of retroactively giving myself reason to prefer that my friends not participate? I didn’t want that! At the time I had genuinely wanted my friends to participate, and I didn’t want to ruin that after the fact!
Of course, I also didn’t want to waste $75 of donations for no reason. Or for dubious decision-theoretic reasons that definitely had not been relevant to the actual behavior of any of the actual people involved.
And so, this has been the convoluted story how I came to donate a particular $75 to GiveDirectly. Usually, I don’t donate to GiveDirectly, because I’m on board with GiveWell’s view that donations to AMF and other top charities are something like ten times as cost-effective as unconditional cash transfers. (See e.g. this funding report.) But I’m a big fan of cash benchmarking—the idea that all global development interventions should be compared to the baseline option of just giving the beneficiaries unconditional cash transfers instead. Similarly, it seemed to me that if money was earmarked for charity, and no one stepped in to make a case for anything else, cash was a sensible default.
I think this approach is potentially useful in a much more common situation: the matching campaign. Matching campaigns are popular, but misleading (see e.g. discussion here, here, and here). The misleadingness could be fixed if the matching donor agreed to set fire to any money that wasn’t used to match other donations, but they typically won’t agree to that (and it wouldn’t be a great look). The idea here is, instead of fire, the matching donor picks a “floor” donation option that they’re okay with their money going to, but not maximally excited about. For best results, the audience of potential donors should largely agree with this assessment, and also the floor option should have some kind of ontological basicness to it, such that it feels reasonable and not insulting to designate it as the “floor”. Obviously I’m describing cash transfers here, but I think other interventions with these properties would work for the same reasons; direct air capture comes to mind.
I’d be interested to hear if something like this has been tried.
Interesting suggestion. I’m not familiar with anyone doing a donation match like this.
It seems like having a default charity for matching money to go to could be counterproductive to the matcher’s goals. E.g. Every.org wanted to get more people to use their platform to donate. But I think many people don’t really find it more valuable for money to get directed to one charity over another. EAs are different in that regard. While we’re certainly not unique in caring which charities money goes to, I think many people might think “Why should I donate when the money is already going to go to charity?” and decide not to participate.
While generally I wouldn’t advise people to do donation matches, would it be good for organizations already running them to make cash transfers the default use of the money if matching donors don’t direct it elsewhere? Maybe. One benefit might be that it just gets people to think more about the value of directing money to one organization versus another, instead of merely thinking that they’re raising more money for a charity of their choice.