Allocating $175M to GiveWell’s Recommendations for 2026

In October 2023, we committed $300 million over three years to GiveWell’s evidence-backed, cost-effective recommendations in global health and development. We recently decided to renew that funding early and increase it, committing an additional $175 million for 2026.

This funding adds to the more than $1 billion in funding we have previously committed to GiveWell’s recommendations. Based on GiveWell’s estimates, we believe that this funding has saved over 100,000 lives.

We don’t usually write in this much detail about individual grants, but given the scale of this commitment, we wanted to share the news about our decision promptly and in some depth.

We’re excited to be renewing our support for GiveWell’s recommendations; we think it’s an outstanding resource and we’re proud to support the high-impact organizations it has identified.

The rest of this post:

• Shares examples of recent GiveWell recommendations we’ve funded.
• Talks through our plans to increase our giving in 2026.
• Explains why we expect that our 2026 funding for GiveWell will roughly equalize GiveWell’s cost-effectiveness margin and our funding bar.
• Lays out other reasons to renew our support for GiveWell’s recommendations.
• Concludes with a brief overview of our plans to continue supporting this work.

Where our funding to GiveWell’s recommendations ultimately goes

A few recent examples of our funding to GiveWell’s recommendations:

• $22 million to the Against Malaria Foundation to support the delivery of insecticide-treated bed nets to prevent malaria in the Democratic Republic of the Congo in 2025–26. GiveWell believes that these nets mostly reach people who otherwise would not have access to them, in areas where malaria is a significant cause of child mortality, and are effective at reducing child mortality related to malaria.
• $3 million to Concern Worldwide to provide acute malnutrition treatment and primary health care in Somalia. The program will support health facilities and mobile clinics that were closed or operating at reduced capacity as a result of aid cuts and establish a buffer stock of therapeutic foods to address potential supply shortages. GiveWell estimates that the grant will avert more than 550 deaths by reducing all-cause mortality among children with severe acute malnutrition and treating common childhood illnesses such as diarrhea and respiratory infections.
• $7.8 million to Evidence Action to expand technical assistance for the Indian government’s prophylactic iron and folic acid (IFA) supplementation program for children to two new states. We’ve previously supported similar technical assistance in other Indian states. GiveWell estimates that this grant will increase IFA coverage by roughly 7 percentage points across tens of millions of targeted children and avert more than 500,000 anemia cases per year.

Our giving is increasing in 2026

We expect our giving to grow significantly next year. The available assets of Good Ventures, our largest funder, are up roughly 50% since the end of 2023, and Good Ventures generally wants to accelerate the pace of its giving. We’ve also seen increased interest from other external funders to partner in the areas where we work. As a result, we are increasing our spending across our portfolio.

We think funding in this range would roughly equalize GiveWell’s bar and ours

At the time of our 2023 allocation to GiveWell, our “bar” for funding in our Global Health and Wellbeing (GHW) work increased to roughly 2,100x (as a result of declining assets, shifting allocations in our portfolio, and finding more cost-effective opportunities over time). With assets and spending up again, we are now decreasing our bar slightly, to roughly 2,000x.

Part of the reason the bar isn’t falling more is that our push since 2019 to identify more cost-effective opportunities has borne fruit, as represented by the Abundance & Growth Fund, Global Growth Fund, and Lead Exposure Action Fund, as well as growing work in existing funds such as global health R&D and global aid policy. Gradually finding more cost-effective opportunities in new areas offsets the need to lower the bar within existing areas as spending rises.

In our previous GiveWell update, we wrote that “in the near term, we expect a small wedge between our estimates for our GHW bar and GiveWell’s bar, possibly on the order of 0-30% but with a wide margin of error,” and expressed uncertainty about how that would affect our future funding for GiveWell recommendations. We also said that we planned to continue evaluating this question.

• Less importantly, GiveWell has raised less funding than we expected based on its early projections. In April 2023, its median projection for funding raised in 2024 from sources other than Coefficient Giving was $421 million; it wound up raising $307 million. When it has less available funding, its marginal opportunity (i.e. the strongest one it can’t fund without us) is more likely to be above our bar.
• More importantly, we’ve significantly revised how we compare GiveWell’s cost-effectiveness units and ours. These updates make GiveWell’s recommendations look 44% more favorable, and suggest that our post-funding margin and GiveWell’s look similar.

This is how our thinking about the comparison between our and GiveWell’s cost-effectiveness units has evolved:

While our current best guess is that the cost-effectiveness of marginal GiveWell and CG grants will be very similar in 2026, this could change over time. As we’ve explained previously, we think that GiveWell’s returns curve is flatter than ours and that our funding for GiveWell should therefore be more sensitive to changes in our funding bar than our funding for other opportunities.

Other reasons to renew support for GiveWell’s recommendations at scale

While equalizing marginal returns was the biggest single driver in our decision, we see a number of other reasons to renew support at significant levels:

• Our partner Good Ventures wants to spend faster, and GiveWell’s recommendations are a robust way to scale spending relatively quickly, even though there are some binding near-term operational constraints that will likely prevent GiveWell from spending all our funding in 2026.
• We’re aiming to be slightly more generous to GiveWell than our calculations would suggest, since we would guess we are mildly biased in favor of our own in-house work relative to that of external partners like GiveWell.
• We see GiveWell’s work as benefiting from relatively strong feedback loops and the ability to learn more over time; this is less true for some other areas we work in, where it is harder to get empirical signals that will change decisions.
• Even if it wasn’t quite as cost-effective, one of us (Alexander) sees additional value in supporting GiveWell’s grants — which are largely safer, more evidence-backed direct service grants — in addition to our riskier, more hits-based grants; this serves to provide a modicum of worldview diversification. Otis doesn’t agree that worldview diversification gives a compelling additional reason for support here; he thinks the benefits are well captured by the ROI comparisons discussed above. We alluded to this consideration in 2018, describing GiveWell as representing “straightforward charity.” With this renewal, we anticipate that GiveWell’s recommendations will likely represent a little over 10% of our 2026 giving.

Plans for future years

We haven’t made a legally binding commitment beyond 2026 at this point, because we want to be able to continue to evaluate the relevant margins between GiveWell’s recommendations and our other work (both of which will change over time). But we expect that we and Good Ventures will continue to be very significant supporters of GiveWell’s recommendations in future years.