Thoughts on My Donation Process

I have some observations and half-baked ideas about my recent donation process. They weren't important enough to include in the main post, but I want to talk about them anyway.

Cross-posted from my website.

On deference

Usually, I defer to the beliefs of other people who have spent more time on an issue than me, or who plausibly have more expertise, or who I just expect to have reasonable beliefs.

While writing my donations post, I made a conscious effort to defer less than usual. Deference might maximize the probability that I make the correct decision, but deference reduces the total amount of reasoning that's happening, which is bad for the group as a whole. I want there to be more reasoning happening.

This is most relevant in my discussion of a few orgs that I disliked, which are also very popular among big EA funders. I'm 99% confident that the big funders have private information about those orgs, so maybe I should defer to them. But I'm also maybe 75% confident that if I had access to that information, it wouldn't materially change my mind. I did anticipate that the private evidence would make me like the orgs a little better, so I updated based on this anticipation and evaluated the orgs a little more favorably than I would have otherwise.

On criticizing orgs

I am not as nice as I'd like to be. I have a habit of accidentally saying mean things that hurt people's feelings.

On the other hand, I think most people are too nice: they hurt others long-term by refusing to give them useful information that's difficult to hear.

(In theory, it's possible to never say unnecessarily mean things, and always say necessary things, but only if you have perfect communication skills. In practice, there's a tradeoff.)

I think it's a good norm that, if you're investigating an org and it opens up to you, you shouldn't take what you learn and use it against the org. I probably wouldn't criticize an org based on private information that it gave me. I did criticize some orgs, but all my criticisms were based on public information.

I think if most people wrote a donation post like mine, they'd self-censor in the interest of niceness and end up leaving out important information. I tried to avoid that, and erred more on the side of being mean (not pointlessly mean, but mean-and-truthful. Or maybe I should say mean-and-accurately-conveying-my-beliefs since I can't promise that the things I said were true).

As with my choice on deference, this was perhaps the wrong choice at an individual level but the right choice at the group level.

I did focus on criticizing organizations and avoided saying negative things about specific people whenever possible.

Donation sizing

I have a donor-advised fund (DAF) that I contributed to when I was earning to give. How much of my DAF money should I donate this year? What's a reasonable spend-down rate?

I've put a lot of thought into how quickly to spend philanthropic resources, including how AI timelines affect the answer. Unfortunately, all that thinking didn't much help me answer the question.

Plus, there are some complications:

• I have some personal savings, which I could choose to donate. Should I count them as part of my donation money?
• I might earn significant income in the future. Right now it looks like I won't, but I might do more earning to give at some point, or I might take a direct-work job that happens to pay well. If I expect to earn more in the future, then I should spend more of my DAF now.
• I didn't donate much money for the last few years. Should I do catch-up donations this year? Or maybe spread out my catch-up donations over the next few years?

I didn't come up with good answers to any of these questions. Ultimately I chose how much to donate based on what felt reasonable.

Diversifying donations as a trade

I had an idea based on this comment by Oliver Habryka. He describes a trade between members of the EA community where some people do object-level work (relinquishing a high-paying job) and others earn money. He argues that when this trade occurs, the people doing object-level work should have some ownership over the funds that earners-to-give have earned.

I spent a while earning to give. So arguably I should donate money to people who started out in a similar position as me but went into direct work instead. Essentially, I should (acausally) trade with altruists who could've earned a lot of money but didn't. And because there are many such people, arguably I should split my donations across many of them instead of only donating to my #1 favorite thing.

But this argument raises some questions. Who exactly was in a "similar position as me"? What about people who aren't members of the EA community, but who are nonetheless doing similarly valuable work? What about people who didn't have the necessary skill set to earn a lot of money, so they never made a choice not to?

I decided not to further pursue this line of reasoning because I couldn't figure out how to make sense of it. I just did the obvious thing of donating to the org(s) that looked most cost-effective on the margin.

Cooperating with the Survival and Flourishing Fund

Should I donate less money to orgs that have received grants from the Survival and Flourishing Fund (SFF)?

I want to be cooperative with SFF. If I donate less to an org that's received SFF funding, that seems uncooperative.

SFF has the S-process, which is a fancy method for allocating donations from a group of value-aligned donors who each want to be the donor of last resort, but who also want to make sure their favored orgs get funded. I could cooperate with SFF by participating in this process.

I asked them if they wanted to add my money to the S-process and they declined, so I consider myself to have officially Cooperated and now I'm allowed to donate less to orgs that received SFF funding. I don't think SFF really cares if its donations trade off against mine because I have much less money than it does.