Our goal with hosting quarterly open threads is to give blog readers an opportunity to publicly raise comments or questions about GiveWell or related topics (in the comments section below). As always, you’re also welcome to email us at info@givewell.org or to request a call with GiveWell staff if you have feedback or questions you’d prefer to discuss privately. We’ll try to respond promptly to questions or comments.
You can view previous open threads here.
Comments
Dear GiveWell Friends,
I was wondering if your algorithm would allow for consideration of the StrongMinds (strongminds.org) charity. They provide cost effective depression treatment programs in Uganda and Kenya, with 80% of participants depression free after completing therapy. While it won’t score nearly as high on “lives saved” as some charities, it certainly alleviates much suffering.
Hi, Brian,
Thank you for your comment, and sorry for the delay in responding!
StrongMinds is the recommended charity of the Happier Lives Institute (HLI), which evaluates giving opportunities using a subjective well-being (SWB) framework—for example, by looking at a program’s impact on participants’ self-reported life satisfaction scores. We recently did our own assessment of HLI’s cost-effectiveness analysis of StrongMinds, using an SWB approach; we estimate that it is about 25% as cost-effective as our top charities, but this is based on several uncertain assumptions, and our estimate could change significantly if we did more research.
We do not typically evaluate programs through an SWB framework. The moral weights we use emphasize averting death, especially in young children, and increasing consumption (a function of increased income). While these moral weights have limitations, we think they capture the most important outcomes we’re looking for in programs, according to input from different groups. We recognize that ours is not the only moral framework out there, and some donors might take a different view that leads them to support StrongMinds instead of our top charities. We also think that, even though our analyses do not explicitly measure SWB effects, our top charities (along with other programs we fund) likely also alleviate suffering by reducing illness and death, and through development effects that could improve children’s lives in the future.
I hope that’s helpful!
Best,
Miranda
Can you please format these threads so that replies are nested under the comment they’re replying to. As it is, reading through these open threads and trying to figure out which comments are replies to which other comments is too difficult
Hi, Henry – thank you for the suggestion! I’ve forwarded it to our tech team.
Best,
Miranda
What do you think is stopping governments from allocating parts of their aid budgets to GiveWell-recommended charities?
Hi, Henry,
Thanks for the question, and apologies that it took me a while to respond!
In fact, governments do allocate some foreign aid funding to the kinds of work done by GiveWell’s top charities, although not always in the same way.
Some examples of foreign aid budgets going to GiveWell-recommended programs:
– High-income countries provide funding for long-lasting insecticide-treated nets, the kind of antimalarial nets distributed in campaigns supported by Against Malaria Foundation, through the Global Fund and (in the US) the President’s Malaria Initiative (PMI). AMF works alongside these funding entities to carry out LLIN campaigns, so its support is going to the same programs. We’ve chosen to support AMF because we’ve found it possible to understand what GiveWell donors’ funds would do through AMF in a way that historically hasn’t been true with larger entities.
– High-income countries also fund Malaria Consortium’s seasonal malaria chemoprevention (SMC) program, as well as other SMC programs, through the Global Fund and PMI.
– Global Affairs Canada, which runs the international development and assistance department of the government of Canada, funds vitamin A supplementation (VAS) programs that work similarly to and in coordination with Helen Keller’s VAS programs.
New Incentives, our other top charity, hasn’t received funding from foreign aid budgets, but it’s also a fairly new program—it’s existed in its current form only since 2017.
We do think foreign aid departments could fill more funding gaps for these areas of work. We speculate that they don’t due to some combination of different values and beliefs; different goals and constituencies to appeal to (e.g., USAID’s work needs congressional authorization, and the average member of Congress is probably not aligned with GiveWell’s approach to finding maximally impactful programs); and different approaches (e.g., GiveWell prioritizes cost-effectiveness in a way that doesn’t seem to be very common in the foreign aid world, but aid agencies also can achieve very high levels of scale and coordination that change the expected impact of a program).
We would be very happy to see more resources from aid agencies going toward GiveWell-recommended programs. One of Open Philanthropy’s focus areas centers on making global aid policy more impactful; you can read more about that work here.
I hope that’s helpful!
Best,
Miranda
Hey GiveWell team, I love your work! I have a couple of questions about the scope of the all grants fund that might help inform other donors and charities I know in the space. It would be super helpful to have your best guesses on the questions below.
1) How does your evaluation bar change based on how new an area is compared to areas that you have historically researched? I understand that GiveWell’s capacity might be limited by research time, and I am unsure how this would affect, for example, the likelihood of comparing two different organizations with similar prima facie cost-effectiveness, but one operates in an area you have already researched, and another in an area you have not previously explored. Do you feel that a charity in a brand new area would have to be 1.1x better to get evaluated, or more like 2x better? I would be keen to hear how this has played out historically, even if it might change going forward.
2) I have a similar question relating to the scale of the organization. If two organizations seemed promising, but one could absorb $20 million a year in donations compared to another that could only handle $1 million a year, what level of increased cost-effectiveness would the smaller organization need to exhibit to be equally likely to be considered?
3) How does GiveWell think about meta funding? I know there have been a small number of historical grants (e.g., to One for the World), but does GiveWell expect to increase the number of grants they are going to give in this area in the next year or two?
4) What is your target ratio of annual funding to incubation program support compared to the classic GiveWell in the next 12-24 months?
5) How many grants have you historically given to policy charities? And what is your best guess for the amount of support you’ll provide to policy charities in the next 12-24 months? What do you look for in a policy charity for it to be funded from the all grants pool?
Hi, Joey – just want to pop in and apologize for the delay, but I’ll have answers for you on the above soon!
Best,
Miranda
Hi, Joey – many apologies for the long delay in getting back to you here! For ease of following along, I’ve copied your questions below in italics and added the answer below each.
1) How does your evaluation bar change based on how new an area is compared to areas that you have historically researched? I understand that GiveWell’s capacity might be limited by research time, and I am unsure how this would affect, for example, the likelihood of comparing two different organizations with similar prima facie cost-effectiveness, but one operates in an area you have already researched, and another in an area you have not previously explored. Do you feel that a charity in a brand new area would have to be 1.1x better to get evaluated, or more like 2x better? I would be keen to hear how this has played out historically, even if it might change going forward.
We are excited to evaluate and make grants to new-to-us programs, particularly because we place a high value on learning about new areas. We would not say that such a program has to look in any measure better than a more familiar one to merit investigation.
Our research team has grown substantially in the past few years, and with it we’ve expanded our capacity for investigating a wider range of programs. Grants to non–top charity interventions have come to make up a significant part of our grantmaking; we directed $190 million in such grants in 2021, vs. $19 million in 2020. With the growth in our research team, we’ve also been able to designate a sub-team that specializes in finding new areas that might yield opportunities for both direct impact and learning (since the more we branch out into new areas, the more we expect to learn). An investigation into a new-to-us and fairly uncertain program that provides a clear path to learning value could be prioritized over something more established if we thought the more uncertain investigation could also “unlock” more cost-effective giving opportunities in the future.
Allocation of staff capacity could be a factor in what we decide to investigate first. If we were capacity-constrained and considering two investigations that were identically promising, but one was more straightforward and the other relatively complex, we’d probably prioritize the straightforward one. But ideally we’d do both, because we now have the “new areas” team to focus solely on this type of work.
2) I have a similar question relating to the scale of the organization. If two organizations seemed promising, but one could absorb $20 million a year in donations compared to another that could only handle $1 million a year, what level of increased cost-effectiveness would the smaller organization need to exhibit to be equally likely to be considered?
If we could only investigate one of these hypothetical programs and we didn’t expect either of them to be able to grow (or expect to learn from them), then the smaller one would need to be 20x as cost-effective. In practice, though, we aim to find programs that we think will provide learning value and/or offer room for growth in the future.
One recent example is a grant we recommended to Suvita, an organization in India that aims to increase childhood vaccine uptake (a formal grant write-up on this is forthcoming). Suvita is very small; we estimate its programs are cost-effective, though not 20x other opportunities. But it’s also well aligned with GiveWell’s orientation toward high impact and transparency, we think it has a lot of room to grow (yielding more cost-effective funding opportunities), and we think we’ll gain valuable information about its programs and similar programs through this grant.
With any promising program, we also want to “rightsize” the amount of time spent investigating to align with the grant size and potential room for more funding. Our officers can make small discretionary grants with minimal time spent evaluating. Outside the discretionary process, we aim to spend less time researching small, straightforward grants, so that we can realize their potential impact while also allocating staff capacity to larger or more complex investigations. These make up a very small portion of our overall grantmaking, and we’re happy to make these small investments that could either be highly cost-effective or generate future opportunities.
We do consider the likelihood that a small organization won’t be able to scale successfully, which would limit its impact. However, for large organizations we’d consider the possibility that they won’t be able to reach the full population in need of the program, similarly curtailing their impact—so we don’t see this as an obstacle unique to smaller organizations.
3) How does GiveWell think about meta funding? I know there have been a small number of historical grants (e.g., to One for the World), but does GiveWell expect to increase the number of grants they are going to give in this area in the next year or two?
We are eager to collaborate with meta funding organizations (defined as organizations that aim to build interest in and fundraise for GiveWell or our recommended programs). While we don’t expect to prioritize investigating potential funding for them in the next couple of years, we also could imagine doing so if our team identified promising opportunities
4) What is your target ratio of annual funding to incubation program support compared to the classic GiveWell in the next 12-24 months?
We don’t have a target ratio in mind—we ideally want to investigate all the promising opportunities that come our way and fund those that meet or exceed our cost-effective bar (currently 10x cash for all programs regardless of top charity status). We don’t have a preference as to how much of our funding goes to top charities vs. other programs.
We projected that we’d direct roughly 75% of our funding to top charities and 25% to other programs in 2022, based largely on the number and size of opportunities in our pipeline at the time. That projection seems to have been roughly correct. So far this year we are estimating a slightly higher proportion of funding to other programs (possibly closer to a 60/40 split than 75/25), but that is very uncertain.
In the near future, we expect that we’ll still direct a majority of funding to top charities and that the ratio will remain in the same ballpark, but that will depend heavily on the outcomes of our investigations into a few very large grant opportunities this year and next.
5) How many grants have you historically given to policy charities? And what is your best guess for the amount of support you’ll provide to policy charities in the next 12-24 months? What do you look for in a policy charity for it to be funded from the all grants pool?
Since 2017, we’ve recommended around $30 million in grants for policy advocacy around public health regulation. (A partial list of grants in this category can be found here. Since that page was published, we’ve also recommended an additional $1.2 million to the Center for Global Development for research and policy work on lead exposure, and $15 million to RESET Alcohol for advocacy to reduce harms from excess alcohol consumption.)
In 2022 and early 2023, we weren’t as focused on policy grantmaking, in part because Open Philanthropy was developing this more fully as a focus area within its global health and well-being work and we wanted to avoid duplicated effort. However, we think it’s likely there are many opportunities in this space that fall outside of Open Philanthropy’s purview and that we’d be interested in looking at. We hope to devote staff time to policy grant investigations this year, and we may recommend such a grant in the micronutrient fortification space, but it is too early to predict how much funding we’ll direct in the near term.
Policy grants are equally eligible for funding from the All Grants Fund as any other type of grant; that is, they must meet our current cost-effectiveness bar.
Hello Give Well Team,
I was recently looking into your job openings and saw that many of them boasted annual salaries well into the six-figure range with many benefits, as well as dozens upon dozens of employees.
Since I am aware that GiveWell does not allocate donations for its own upkeep and funding unless the donor choose to do so, I am made to wonder how GiveWell supports itself so robustly. Is this information publicly accessible?
I can easily see how something like this could be misconstrued as a criticism of all EA-type organizations (and charities in general), so I thought I would ask.
Thank you for your work.
Hi, Garrett,
Thank you for your question! GiveWell funds salaries and benefits for staff out of our operations budget, which is supplied by donations to our unrestricted fund. The best source of information for how we use our unrestricted funding is our annual metrics report, the most recently published of which is for 2021.
In 2021, we received $110 million in unrestricted donations, out of a total of $595 million in funds raised. Of our unrestricted funding, we spent $11 million on our operations, $6 million of which went toward staff and contractors; the remainder was granted out in accordance with our excess assets policy and our single-donor cap policy. (See the “GiveWell’s operations” section starting on page 10 of the 2021 metrics report.) Total spending on staff and benefits in 2021 was about 0.01% of our total funds raised.
Our staff has been growing significantly to catch up with the rapid expansion of our funds raised between 2019 and 2021. We see the growth of our research team especially as essential to our ability to fulfill our mission by directing these funds impactfully.
We set compensation levels for our staff based on market data representing what a typical employee with the skills and background we look for could earn elsewhere, as well as cost of living in employees’ geographic locations and their demonstrated performance over time. Our salaries are competitive with those of peer organizations like private foundations, academic institutions/think tanks, and the World Bank.
I hope that’s helpful—please let me know if you have any further questions!
Best,
Miranda
Hi, Garrett,
Sorry, I made a mistake in my previous answer! In 2021 we spent about 1% of our total funds raised on staff and contractors, not 0.01% ($6 million out of $595 million). Apologies for any resulting confusion!
Thanks,
Miranda
In a Twitter conversation, Anthony Kalulu the creator of Uganda Community Farm has criticized GiveWell for not giving any recommendations of charities that were created by people in the Global South (in the belief that people located in that region are in the best position to know what interventions will be helpful). Although I do not necessarily endorse this criticism, it would be helpful to know GiveWell’s response so that I could pass it on to Anthony.
Hi Rupert,
Thank you for reaching out!
GiveWell is dedicated to finding the charities that save or improve lives the most per dollar. We recommend a few outstanding charities to help donors make a significant impact with their gifts, and we direct funding to many other excellent organizations and programs. You can also take a look at all grants that have been made or recommended by GiveWell since 2014 here.
We specifically look for interventions that are evidence-backed, cost-effective, transparent, and underfunded, rather than applying specific criteria around where the charity may be headquartered. However, we certainly consider and have funded organizations such as IRD Global that are founded and led in the countries where they work.
That said, we certainly do not believe that ours is the single correct or best approach to giving. We appreciate your feedback, as it helps us to ensure we are addressing any concerns, and we understand that other donors may have different priorities that they find important. Our hope is to be explicit and transparent about our framework rather than to assert that it is best.
Please feel free to reach out with any other questions.
Best,
Miranda
Thank you for your commitment to transparency, and your thorough responses to public questions. While death aversion is necessary to ensure many people live, it does not ensure they are living in healthy living conditions. Why are nonprofits who focus on developing basic infrastructure (such as access to clean water and sanitation) omitted from GiveWell’s list of recommended charities?
Hi, Max,
Thank you for your question!
We fund a number of programs outside of our top charities, many of which provide life-improving benefits in addition to, or instead of, saving lives. Donors who want to support these programs can do so by giving to the All Grants Fund; you can see a partial list of all grants we’ve recommended (for both top charities and other programs) since 2014 here. For example, we’ve recently directed funding to malnutrition treatment, micronutrient fortification, deworming, syphilis screening and treatment, kangaroo mother care for low-birthweight infants, and trailbridge building.
Water treatment programs have been a significant area of our research in the last couple of years. We have funded clean water access through grants to Evidence Action for chlorine dispensers and in-line chlorination, and have also funded research into vouchers for chlorine. We’ve focused on programs that treat water, rather than those that build new water infrastructure, because we’ve learned in the course of our research that water treatment is relatively underfunded compared with water infrastructure. We haven’t added any water treatment programs to our top charity list, because we haven’t yet determined that they meet the additional criteria we’ve established for top charities.
I hope that’s helpful!
Best,
Miranda
extremely helpful – many thanks!
Do you have any plans to research the efficacy of releasing genetically modified mosquitoes to reduce malaria disease burden?
Hi, Dylan,
Thanks for your question! We’re not actively planning to look into this any time soon, though it’s a possibility for further in the future. However, Open Philanthropy, with which we work closely, has funded some work on gene drives that could help eliminate malaria in sub-Saharan Africa, which you may be interested in reading about.
Best,
Miranda
You define cost-effectiveness as “saving or improving lives as much as possible for as little money as possible”. Does this apply only to the lives of people who are currently alive, or also to lives of future generations? What is the justification for your choice?
Hi, Dan,
Thanks for your question! Our analyses don’t typically model the effects of programs on future generations, even when they may have such an effect. We think GiveWell’s research approach is strongest at identifying opportunities to help people on relatively short time horizons, with relatively high levels of confidence. We would guess that GiveWell’s marginal impact on the long-term future might not be as significant as on the health and well-being of people in the near term, so we plan to continue focusing on making an impact over shorter time horizons.
We do think many of the programs we support could have a positive impact on future generations. For instance, antimalarial nets and chemoprevention can be part of a country’s strategy to eliminate malaria, and incentives for vaccination against other infectious diseases could contribute to eliminating those diseases in parts of the world. We have also funded grants in public health regulation, e.g., for advocacy work to reduce lead exposure, that we would expect to benefit lives in the further future by changing laws; we historically modeled a longer time horizon (our default is 10 years) for the benefits of these grants.
I hope that’s helpful!
Best,
Miranda
What is the GiveWell position on programs to expand access to contraceptives in poor areas? I only found this link from 2017 indicating ” This program appears promising, and we plan to continue our investigation.”
Is there a followup?
https://www.givewell.org/international/technical/programs/sayana-press
Hi, Tim,
Thanks for your question, and sorry for the delay in getting back to you!
We have briefly looked into the effectiveness of several contraceptives that could be useful for expanding contraceptive access, although we have not yet conducted complete assessments of the evidence for these, or looked into specific programs that provide them. These include Sayana Press, contraceptive implants, and intrauterine devices (IUDs). We have substantial uncertainty over the value of different outcomes related to family planning, which has made it challenging to model the benefit of programs supporting use of these tools. We are continuing to explore this area.
As we do more research around how to weigh these different outcomes, we are considering some limited potential investments in other family planning programs. In March of this year, we recommended a $500,000 grant to Family Empowerment Media (FEM), which broadcasts informational radio content about family planning in Nigeria; specifically, this grant is intended to support a randomized controlled trial (RCT) that FEM is planning of its program. A write-up about this grant is forthcoming. We believe that FEM’s program might be very cost-effective, but that an RCT could help us refine some inputs into our analysis before we make a decision on funding the program.
I hope that’s helpful!
Best,
Miranda
Comments are closed.