I’ve donated about $150,000 over the past couple years. Here are some of the many (what I believe to be) mistakes in my past giving:
Donating to multiple cause areas. When I first started getting into philosophy more seriously, I adopted a vegan lifestyle and started identifying as EA within only a few weeks of each other. Deciding on my donation allocations across cause areas was painful, as I assign positive moral weights to both humans and animals and they might even be close in intrinsic value. I felt the urge to apologize to my vegan, non AI-worrier friends for increasing my ratio of AI safety donations to animal welfare donations, and my non-vegan, non-EA friends and family thought that donating to animals over humans was crazy. Now my view is something like: donations to AI safety are probably orders of magnitude more effective than to animal welfare or global health + development, so I should (and do) allocate 100% to AI safety.
Donating to multiple opportunities within the same cause area. Back in my early EA global health + development days, I found and still find the narrative of “some organizations are 100x more effective than others” pretty compelling, but I internally categorized orgs into two buckets: high EV and low EV. I viewed GiveWell-recommended organizations as broadly ‘High EV,’ assuming that even if their point estimates differed, their credence intervals overlapped sufficiently to render the choice between them negligible. This might even be true! However, I do not believe this to generalize to animal welfare and AI safety. Now I’ve come full circle in a way, and believe that actually, some things are multiple times (or even orders of magnitude) higher EV than other things, and have chosen to shut up and multiply. If you are a smaller donor, it is unlikely that your donation will sufficiently saturate a donation opportunity such that your nth dollar should go elsewhere.
Donating to opportunities that major organizations recommend/fund publicly. Major organizations may face constraints that individual donors do not. Non-profits are limited in the political activity they can engage in. Large funders may face reputational constraints that make certain grantees a poor fit. For instance, CG has noted that right-of-center AI policy groups may not be a good match for their main funder despite potentially doing valuable work, and certain cause areas may be too weird for certain funders.
Donating at the end of the year. Major evaluators often post their public recommendations at the end of November because philanthropic activity spikes in December due to holidays and the end of the tax season. The best donation opportunities do not only appear in December! If I’m donating to 501(c)(3)s and trying to optimize taxes, I use a DAF so that I can donate in a month other than December. But also, the tax status of an organization is not a proxy for impact. For example, in the US, donating to 501(c)(3)s and 501(c)(4)s may provide tax benefits. Assuming you would donate the funds saved on taxes, it still may be higher EV to donate to non-501(c)(3)/501(c)(4) opportunities and just take the tax hit. Additionally, time discounting may be steep enough such that you should make sacrifices (tax or otherwise) to donate now rather than later.
I used to donate mid-year for the reasons you gave. The last couple years I donated at the end of the year because the EA Forum was running a donation election in early December, and I wanted to publish my “where I’m donating” post shortly before the donation election, and I don’t want to donate until after I’ve published the post. But perhaps syncing with the donation election is less important and I should publish and donate mid-year instead?
I’ve donated about $150,000 over the past couple years. Here are some of the many (what I believe to be) mistakes in my past giving:
Donating to multiple cause areas. When I first started getting into philosophy more seriously, I adopted a vegan lifestyle and started identifying as EA within only a few weeks of each other. Deciding on my donation allocations across cause areas was painful, as I assign positive moral weights to both humans and animals and they might even be close in intrinsic value. I felt the urge to apologize to my vegan, non AI-worrier friends for increasing my ratio of AI safety donations to animal welfare donations, and my non-vegan, non-EA friends and family thought that donating to animals over humans was crazy. Now my view is something like: donations to AI safety are probably orders of magnitude more effective than to animal welfare or global health + development, so I should (and do) allocate 100% to AI safety.
Donating to multiple opportunities within the same cause area. Back in my early EA global health + development days, I found and still find the narrative of “some organizations are 100x more effective than others” pretty compelling, but I internally categorized orgs into two buckets: high EV and low EV. I viewed GiveWell-recommended organizations as broadly ‘High EV,’ assuming that even if their point estimates differed, their credence intervals overlapped sufficiently to render the choice between them negligible. This might even be true! However, I do not believe this to generalize to animal welfare and AI safety. Now I’ve come full circle in a way, and believe that actually, some things are multiple times (or even orders of magnitude) higher EV than other things, and have chosen to shut up and multiply. If you are a smaller donor, it is unlikely that your donation will sufficiently saturate a donation opportunity such that your nth dollar should go elsewhere.
Donating to opportunities that major organizations recommend/fund publicly. Major organizations may face constraints that individual donors do not. Non-profits are limited in the political activity they can engage in. Large funders may face reputational constraints that make certain grantees a poor fit. For instance, CG has noted that right-of-center AI policy groups may not be a good match for their main funder despite potentially doing valuable work, and certain cause areas may be too weird for certain funders.
Donating at the end of the year. Major evaluators often post their public recommendations at the end of November because philanthropic activity spikes in December due to holidays and the end of the tax season. The best donation opportunities do not only appear in December! If I’m donating to 501(c)(3)s and trying to optimize taxes, I use a DAF so that I can donate in a month other than December. But also, the tax status of an organization is not a proxy for impact. For example, in the US, donating to 501(c)(3)s and 501(c)(4)s may provide tax benefits. Assuming you would donate the funds saved on taxes, it still may be higher EV to donate to non-501(c)(3)/501(c)(4) opportunities and just take the tax hit. Additionally, time discounting may be steep enough such that you should make sacrifices (tax or otherwise) to donate now rather than later.
I used to donate mid-year for the reasons you gave. The last couple years I donated at the end of the year because the EA Forum was running a donation election in early December, and I wanted to publish my “where I’m donating” post shortly before the donation election, and I don’t want to donate until after I’ve published the post. But perhaps syncing with the donation election is less important and I should publish and donate mid-year instead?