For this post, some Open Phil staff members wrote up the thinking behind their personal donations for the year. Staff are listed in order of their start dates.
You can click the below links to jump to a staff member’s entry:
- Holden Karnofsky
- Alexander Berger
- Nick Beckstead
- Helen Toner
- Claire Zabel
- Lewis Bollard
- Ajeya Cotra
- Morgan Davis
- Michael Levine
I front-loaded my giving last year, and consistent with that, I am not giving this year.
This year, I’m planning to follow the same allocation I did last year:
- 80% of my contributions to GiveWell for regranting at its discretion
- 5% to GiveWell for operations. I place huge value on the top charities research, and I think it’s totally appropriate for me to contribute to pay for it. This decision probably doesn’t cost me anything because I think marginal operations contributions are fungible with support to top charities, but I hope that I can save my former colleagues some time on fundraising efforts and promote the idea that it’s a good choice for donors who use GiveWell’s research to direct some portion of their giving to support it.
- 5% to GiveDirectly. As I’ve said for the last two years, “I continue to feel that they are a uniquely outstanding organization and add a huge amount of value by serving as a benchmark against which other organizations and interventions can be compared… I don’t think that my contribution will add as much value with GiveDirectly as it would elsewhere, but if I didn’t give significantly to GiveDirectly this year, I’d want to again within the next few years to renew my claim to being a ‘supporter’ and because I find them a particularly valuable organization to be able to discuss when making the case for giving a lot.”
- 10% to farm animal welfare groups recommended by Lewis. I remain uncertain about how to weight animal organizations against those focused on humans, but continue to believe that they are worthy of moral concern and that Lewis has identified groups with a strong potential to cost-effectively alleviate suffering.
I wasn’t involved in the GiveWell top charity process at all this year, so after the recommendations were made, I followed the public directions and made a copy of the cost-effectiveness spreadsheet and filled in the median of staff values for empirical premises and my own current best guesses for the moral weights. Relative to cash, I ended up with pretty similar estimates to the median staff members for most charities, but was ~50% more optimistic about bednet distribution. This is driven by an evolution in my thinking towards a somewhat unusual view of population ethics and a single paper I haven’t evaluated thoroughly on the empirical consequences of bednet distributions for population growth (see David Roodman’s older and infinitely more thorough report on related subjects here), so I don’t put much weight on my own view here. Given that GiveWell’s regranting is likely to result in a majority of donated funds going to AMF, which distributes insecticide-treated bednets, and I put little stock in my own views here anyway, I feel good about deferring to their expertise and judgment.
I spent a while thinking about the ramifications of the tax bill currently being considered in Congress for whether I should give more this year vs next year. While things are still in flux, the contemplated change in my marginal tax rate is small enough that the effective marginal cost of my charitable giving is likely to be relatively unchanged this year vs next. The main implication I see for people who are in situations similar to mine (unmarried, live in a higher-tax state like CA & NY, above average income, give significantly to charity) is that the elimination of the state and local income tax deduction and the increase to the standard deduction would significantly increase the returns to bundling contributions from multiple years into one year. So in giving next year, I may decide to write a check in early January 2019 instead of late December 2018, but otherwise I don’t see much in the tax bill that would affect my giving. For people who are in much higher tax brackets, there may be somewhat more incentive to give this year vs next, though again not by a huge amount (looks like the biggest cuts in marginal rates are 4.6% for married couples making more than $480K). For single people who donate less than $12K per year (the new standard deduction; it is $24K for married couples) and do not otherwise itemize, the returns to either bundling or pulling donations forward this year both increase a lot, and giving more this year and less in future years might be a stronger consideration than it is for me. [Please note that this was drafted prior to the release of the conference committee version of the bill, which as of 12/13 has been reported to preserve the state and local income tax deduction up to $10K, which would reduce the returns to bundling.]
As in past years, I considered and ultimately decided against devoting my annual giving to one of the organizations we’ve come across at Open Phil. I continue to spend the bulk of my working hours on Open Phil’s policy efforts, and see supporting other organizations with my charitable contributions as an attractive form of diversification (even though I don’t generally think it’s useful to diversify in charity). That said, some of the evolution in my thinking on population ethics has made me more sympathetic to efforts focused on the far future and global catastrophic risks, and if there were specific opportunities there that were funding constrained, I think I would consider allocating more of my personal giving there in the future.
A final note: as I’ve read other staff members’ plans for their giving, and the team at GiveWell’s, I’m reminded again of how lucky I am to work with (and in the latter case, benefit from the work of) such a thoughtful and dedicated group of people. I’m really grateful for all of the work that they do. Thanks folks.
I am planning to give to my personal DAF this year. I expect to use the funds for purposes broadly similar to the EA Funds that I manage. My views about how it is best to give to are generally similar to the views I expressed last year, though my opinions about “room for more funding” considerations are different. I plan to briefly note some updates to my views in Open Phil’s upcoming “suggestions for individual donors” post.
This year, I gave the large bulk of my donations to a new project being started up by a couple of friends. (The project is currently non-public, so I won’t give more details here.) I generally think that funding early-stage projects where you have inside information, e.g. by knowing the people involved, is one of the very best ways for small donors to give, so I was excited to come across this project. In this case, I know the founders pretty well and have confidence that they share my values, that they can competently execute the project, and that receiving additional funding quickly was helpful to them, so I saw this as an unusually good opportunity for me as an individual.
I have a small amount of my donations budget remaining, and am still deciding what to do with it. By default I will most likely donate it to GiveWell to be allocated to top charities at their discretion.
I don’t plan to make any large donations this year, because, like Holden, I front-loaded my giving last year. I’m a small donor, but I expect that investigating opportunities to give and considering my choices carefully will substantially increase the impact of my giving, but the time spent investigating also has a substantial opportunity cost. Because of that, I prefer making fewer larger donations to a greater number of smaller ones.
I know of some other strategies for converting many small donations into fewer larger ones, such as contributing to donor lotteries (a new one has recently opened on EA Funds: Donor Lottery) or DAFs such as my colleague Nick’s EA Funds. Since I would also like to save time by bundling decisions about my giving process, I plan to wait until next year to investigate whether I should switch my process of donation consolidation from alternating-year donations to another method.
I expect the bulk of my donations to go to projects aimed at improving expected outcomes of the long-term far future.
I’m giving exclusively to farm animal welfare this year because I think it continues to be especially neglected relative to its importance and tractability — which I think means there are more “low hanging fruit” for donors to take advantage of. (This has improved since Open Phil and others have entered the field: two years ago I estimated that less than $30M was spent on farm animal advocacy worldwide; I estimate it’s now closer to $75M. But this is still tiny compared to the roughly 135B farmed mammals, birds, and fish potentially suffering at any time — it’s less than a tenth of a cent per animal.)
My goal in my recommendations for Open Phil grants and EA Animal Welfare Funds disbursements is to maximize the marginal impact of each dollar in reducing animal suffering. But I’m less sure this should be the goal of my personal giving, given it’s likely not worth the time to try to maximize the impact of my small donations when I could focus on much larger Open Phil grants. So I decided this year to use my personal giving in support of my Open Phil grants: primarily as a tool to learn, but also to develop better relationships with potential grantees and to attend events where I could have useful conversations with donors and advocates.
As such, I’ve donated to 15 groups so far this year, including promising new ones (the EA Funds, the Sentience Institute, Encompass, and others) and strong established ones (The Humane League, the Good Food Institute, Mercy for Animals, Compassion in World Farming USA, Animal Equality, and others). I doubt that splitting donations this widely is a good strategy for most individual donors. It made sense for my goal of learning and furthering relationships.
I plan to devote more time in the coming weeks to the topic of the best animal welfare individual giving opportunities. I’ll be sharing my more informed thoughts in our Open Phil individual giving recommendations blog and a newsletter I’ll be publishing on the same topic (you can sign up here). If you’re reading this in pursuit of inspiration about where to give to support farm animals, you can either wait for those recommendations or donate to Animal Charity Evaluators’ top charities or EA Animal Welfare Funds (which I manage).
I am planning to participate in a donor lottery run by the Center for Effective Altruism this year. Regardless of your choice of cause area or opinions about giving now vs giving later, I think that giving to a donor lottery should be a strong default choice for small donors, for largely the same reasons that many staff members mentioned last year:
- It saves thinking time: for donors giving at my scale, it is generally not an efficient use of time and attention to carefully optimize their gift. With the lottery, donors only need to think carefully about where they should give if they win a substantial amount of money; at those scales, it generally is worthwhile to put effort into optimizing one’s donation.
- Moving to larger scales of giving can create qualitatively different opportunities: at these scales, it may be possible to provide seed funding for a new nonprofit or startup, or allow a charity to make a specific new hire or start a specific sub-program, or make it possible to give somewhere which would normally not be willing to accept small donors’ money.
This year in particular, I happened to come across a particularly promising and unusual opportunity (to give to the early stage organization mentioned by Helen). I don’t expect to change my mind if I win the lottery, and I don’t believe I gain any qualitative advantage from scaling up my giving. Nonetheless, I am participating in the lottery anyway to get into the habit of making it my default mode of giving, and to signal my strong support for the institution.
This year my donation will all go to Effective Altruism Funds (EA Funds), with a split of 5% to Animal Welfare, 5% to Global Development, 15% to EA Community and 75% to the Long-Term Future Fund. EA Funds is sort of like a mutual fund for charitable dollars - it allows me to make a single donation and choose how it will be allocated across different cause areas, and fund managers in each cause area will then regrant the money. I’m really excited to use EA Funds, since it gives me a quick and hassle-free way to capture the gains of specialization and pooling resources.
Before deciding to give via these funds, I briefly considered giving to Berkeley Existential Risk Initiative (BERI), which is a new org that provides administrative and operations support to individuals and orgs working on existential risk projects. In my work as Open Phil’s grants manager, I’ve become familiar with how challenging it can be to navigate university bureaucracies and get internal approval for expenditures, even if the purchase might be a very reasonable expense for a project. For instance, a researcher might be unable to hire an event coordinator for a conference or a web designer for a website launch. BERI is designed to help provide more flexible funds and support for these researchers, and as a new nonprofit, doesn’t yet have a diversified and established donor base.
However, I think my impulse to consider directly funding this single org is coming from the same kind of reasoning that makes individuals want to play the stock market themselves, when they’d often do better with a mutual fund. While I do happen to know more about this particular kind of issue, I think that is an accident of my current role, rather than representative of a true, pressing, disproportionate need or opportunity within the cause area. What I really care about is supporting EA and long-term future cause areas in general, and my colleague Nick, who manages these two funds, knows far more about the track record, funding gap, current work quality, and planned projects for not just BERI but many other promising orgs - such as Future of Humanity Institute, Machine Intelligence Research Institute, Centre for Effective Altruism, 80,000 Hours, Centre for the Study of Existential Risk, Global Priorities Institute, Center for Applied Rationality, etc.
Since I broadly trust Nick’s judgment and he has a clearer and more developed view into this ecosystem than I do, then I think he’s in a better position than me to identify the most exciting opportunities and give consolidated donations (which are more useful to orgs trying to plan and execute ambitious projects). That said, if I had special knowledge into an unusually promising opportunity or if I felt my values or judgment were significantly unaligned with a particular fund manager, then I’d be less motivated to give via a fund.
As for why I chose to allocate 75% of my donation to the Long-Term Future fund, for my value set, enabling a positive long term future scores very high on neglectedness and importance. Although this cause area seems less tractable and has murkier feedback loops than animal welfare or global development, nonetheless, I think there is clearly more reasonable activity the world could engage in to mitigate existential risks and increase our odds of achieving a positive long term future, and I’m excited to support this work. Given that, I considered giving my donation exclusively to the Long-term Future fund, but decided instead to give some across all the funds. I’m giving 15% to the EA Community fund because increasing the number and efficacy of smart, motivated people working on the world’s most pressing problems would be (it seem pretty obvious to say) really great, and I think there’s reason to believe the EA community is doing this, so I’d like to support the growth and health of that community.
I’m giving 5% donations each to the Animal Welfare and Global Development funds for a couple different reasons. One is that to my mind these are clearly pressing and valuable causes; if I wasn’t concerned about the long term future or generally improving the number and efficacy of people using reason and evidence to substantially improve the world, I’m pretty confident I would be focusing my donations directly on these causes. Given that, I feel more comfortable keeping at least some investment in these projects - after all, I might be wrong. In addition to providing a small hedge to my portfolio, I think this donation may also buy me some mental flexibility. I agree with Claire’s post that giving small amounts to different cause areas might be a helpful precaution against getting overly invested in a particular worldview. If I was working in a different job, I might have tried to diversify my giving a bit more because of the “might be wrong” concern. However, I’m very diversified in my work life, since I provide support for projects across all of Open Phil’s cause areas, so I’d like to be more focused in my personal giving.
I am planning to give the majority of my charitable contributions this year to GiveWell’s top charities, with 70% going to the Against Malaria Foundation (per GiveWell’s recommendation) and the remaining 30% to GiveDirectly. I chose to deviate from GiveWell’s recommended split in part because (as others have explained more eloquently in past years) I see value in having direct cash transfers serve as a benchmark for other interventions, and in part because I want to signal my excitement about GiveDirectly’s new universal basic income project, even if my contribution will not meaningfully advance that specific work.
I believe UBI could have significant implications for improved subjective well-being for very many future people,1 and think contributing to GiveDirectly this year offers a unique opportunity to get in on the “ground floor” of that initiative’s future impact while simultaneously having a measurable, evidence-backed, low-risk positive impact on the poorest people alive today.
I also made a number of small donations throughout the year to some of the organizations mentioned in our February blog post Giving Suggestions Pertaining to Recent Executive Actions. I considered whether to support one or more of the many wonderful organizations working in Open Phil’s focus areas, but decided that my limited impact as a small donor made this less attractive. I might well update this thinking in future years.
- 1. I understand the caveats about limited applicability of lessons from UBI projects to help the global poor today to potential future UBI projects to mitigate the impacts of automation on employment.