On running fundraisers for weird charities

By DanielFilan @ 2016-01-14T19:35 (+14)

It's reasonably common in the EA community for people to run Christmas or birthday fundraisers for EA charities. Usually, these seem to be for charities like AMF – partially because Charity Science has great infrastructure for running Christmas/birthday fundraisers for GiveWell recommended charities, and partially because a large fraction of the EA community believes that these are the best charities to fundraise for. In addition to these reasons, I suspect one factor is that many people believe that their friends and family would be less willing to donate to “weirder” charities, all else being equal. EA Forum user Lila expresses this concern in a comment:

I wonder how sensitive these fundraisers are to the identity of the organization. I would rather fundraise for GiveWell or CEA than deworming (obviously MIRI, FHI, etc. would be infeasible), but I imagine that would be a less popular choice for donations, since those organizations don't read as "charity" in the same way.

I personally suspect that Raising for Effective Giving (or REG), a meta-charity that convinces poker players to give to effective charities, is the most promising target for donations, for essentially the reasons laid out by Michael Dickens in his cause prioritisation post. Because of this, I decided to run a Christmas fundraiser for REG, and wanted to figure out whether people were less inclined to donate to it because of its weirdness. For my fundraiser, I emailed (or sometimes FB-messaged) 42 acquaintances a message like this:

Hi [redacted],

It's been ages since we've caught up - how are things going for you?
Would you be almost graduating from your psych degree? If so, any idea
what's next?
My life is going well - in particular, I managed to get a university
medal! I've also got an internship at the Future of Humanity Institute
at Oxford, which should start in late January and go for three months.
My PhD applications have been sent out, and I'm nervously waiting to see
where I get accepted.
I'm actually in Sydney for a few days (including right now), but
unfortunately I need to spend time with my family and organise stuff for
leaving the country on Monday (I'm visiting American family and then
chilling with friends in SF), so I don't have time to catch up in
person, sorry :(
This Christmas, I've decided to run a fundraiser for a charity called
Raising for Effective Giving (REG). REG is an organisation that
convinces professional poker players to donate a portion of their
winnings to the most effective charities. I'm very excited by their work
- not only are they causing a large amount of money to be moved to
charity (in fact, it's about $10 moved for every $1 donated to REG), but
they are also getting this money moved to charities that most
effectively help others. Some charities can do hundreds of times more
good with donations than others, so getting these donations to more
effective charities is a huge deal. They've been pretty successful in
the poker world, and are expanding to other fields, such as gaming,
finance, and daily fantasy sports.
If you want to support REG, there are two options. First, if you're up
for giving more than $70, just make a donation at
http://reg-charity.org/donate/ to the 'Effective Altruism Foundation',
the organisation behind REG (note that you can't donate in AUD, only
USD, GBP, EUR, and CHF). If you'd like to give a smaller amount, let me
know, and I'll make a list of these donations in the comments of this post:
https://www.facebook.com/danielfilan/posts/1020515235879. At the end
of the fundraiser (i.e. Boxing day, or as soon after as I can get around
to it), I'll donate the total amount to REG, and get you to transfer me
the money at your convenience.
Obviously, it's totally fine if you don't want to give. At any rate, I'd
really like to hear how things have been going for you, and hope that
you have a merry Christmas and a happy new year.
-Daniel

It took approximately 10 hours to send out all of these emails, although I didn't carefully track the time I spent on it.

Now, there were factors that worked for and against my fundraiser. Firstly, I was contacting relatively few people, so the average person that I contacted was closer to me than the average person that others may contact in their fundraisers. I also raised for one of the least weird 'weird' charities: in order to think that donating to REG is a good idea, you don't have to accept weird ideas like “it's incredibly important to support the EA movement” or “we should be incredibly concerned by existential risk from superintelligent AI”. That being said, REG is still weirder than AMF. It is also harder to fundraise for than AMF: REG doesn't really have good infrastructure for running Christmas/birthday fundraisers, and it's actually non-trivial for people to figure out how to give them money. Finally, I suspect that my email was long and not optimally written, which could make people less likely to donate.

In total, I got AU$800 = US$590 in donations from 10 people. Most of the money came from a few people who I would call semi-EAs – my impression was that they sort of knew about EA ideas, but that they weren't part of the community (I didn't contact my EA friends, figuring that they would be donating effectively anyway). This comes in at US$59/hr, which is more than I would have made otherwise.

The other question is whether I could have gotten more money by fundraising for a more normal-seeming charity like AMF. The answer appears to be no: out of the two fundraisers that I found people writing about on the EA forum, they both raised US$14.3/person contacted, while I raised US$14.0/person contacted¹. I'm not including donation matching in these numbers, since I want to measure people's willingness to donate, and I'm unsure how counterfactually valid the donation matching was². 

Anecdotally, few seemed to worry about REG seeming weird, and some responses indicated that they thought that it was an interesting idea. One person was concerned that REG could make the gambling industry look good and therefore cause more people to be involved in problem gambling, but that was the only weirdness-related concern (and in fact the only concern at all) that I heard.

I think that this is pretty decent evidence against a strong weirdness effect on people's willingness to donate to fundraisers if you have a friend group like mine – and my guess is that my set of friends and family is not that different from the average EA's. It is also weak evidence against a weak effect, perhaps on the order of a 10% reduction in money raised per person contacted, but I wouldn't rule this out. I'm not sure whether this generalises to weirder weird charities such as CEA or MIRI, but I would update in that direction.

My conclusion would be this: if you think that donations to a weird charity are at least twice as valuable as donations to the best normal charity, and you want to run a birthday/Christmas fundraiser, it seems worth it to fundraise for the weird charity despite the possibility of eliciting lower donations per person contacted. That being said, if you think that donations to weird charities are very close to twice as valuable, it might be worth having the fundraiser for the normal charity if you can get counterfactually-valid matching funds for the normal charity but not the weird charity. Finally, I think that it would be worth having birthday/Christmas fundraisers for weirder weird charities to see if weirdness effects kick in for those.

Footnotes:

1. These numbers are weirdly close. This could just be due to nominal effects: maybe people tend to give $20, and Australian dollars are pretty much the same as Canadian dollars. Still, I'm pretty surprised by this, and would be interested in data from other fundraisers.

2. Also, and more boringly, these numbers are calculated using exchange rates at the time that the respective fundraisers were run, but were hard to compare since they were in different currencies that have different exchange rates at different times. I also didn't adjust for PPP, which might make a difference.