Donors collectively gave $11.7 million to "Grants to recommended charities at GiveWell's discretion," which we recently renamed the "Maximum Impact Fund," in the first half of 2020. As always, we grant these funds where we believe they will have the most impact. In addition to this $11.7 million, we disbursed the following large, individual donations as part of our flexible funding:
- $3.6 million was from a donor who provided GiveWell with unrestricted funding. Unlike "Grants to recommended charities" funding, unrestricted funding may be used to support our operations. We chose to grant this portion of the donor's gift because GiveWell will only fund up to 20% of our operations from a single donor, and this donor had exceeded that amount. We included this funding in our $15.3 million grant to AMF.
- $8 million was from a donor who gave to GiveWell to allocate to our top charities but wanted the funding to be distributed sooner than we planned to allocate our "Grants to recommended charities" funding. We decided to grant these funds to Malaria Consortium's seasonal malaria chemoprevention program. We discuss this in the "Other options considered" section of the post.
 Our most recent estimate is that AMF averts a death for between $4,000 and $5,000. We use the more conservative estimate here, as we think it is appropriate to be conservative when sharing impact estimates. $15.3 million divided by $5,000 per life saved = 3,000 lives. GiveWell: 2020 cost-effectiveness analysis — version 1, "AMF" tab, cell B219.
 For information on AMF's work, see Against Malaria Foundation: AMF activity update: March to May 2020 (archive).
 In November 2019, we wrote that we were not planning to make a large grant to AMF, beyond the incentive funding each of our top charities receives for engaging with our process, until we resolved three questions: "We decided not to recommend filling AMF's room for more funding beyond the $2.5 million incentive grant. Over the next few months, we plan to revisit the possibility of making a large grant to AMF. Our decision will rely on what 2020 distribution agreements AMF signs in the near future (if any); how well we are able to answer outstanding questions we have about AMF's monitoring; and, if available, results from AMF's 2019 distributions in DRC, where it is likely to commit marginal funding." GiveWell: Recommendation to Open Philanthropy for Grants to Top Charities, November 2019.
At the time we decided to allocate additional funds to AMF, monitoring from AMF's 2019 distributions in DRC was not yet available. AMF was planning to conduct this monitoring in August 2020, and we did not expect to see results until after our grant decision was made. This is a limitation in our understanding of AMF's planned work.
 GiveWell's room for more funding analysis for AMF , "Funding commitments," cell B5.
 The total in uncommitted funding was $4,157,598. GiveWell's room for more funding analysis for AMF [June 2020], "Available and expected funding (overview)," cell B11.
 AMF told us that it would prioritize the following distributions with the grant funding:
- Two provinces in DRC: Kasai and Kasai Oriental ($13.7 million). The distribution is scheduled for January and February 2022, which would be 36 to 38 months since nets were last distributed in those locations. AMF believes that nets should ideally be distributed every 30 months because of evidence showing that net coverage is low after this point. This is consistent with the evidence we've seen from smaller-scale AMF distributions in DRC in 2014 and 2016 and from a previous, non-AMF distribution in DRC:
- A follow-up study of AMF nets (for a distribution in DRC that ended in June 2016) found that after 24 months, 31% of nets remained in the households in usable condition. Summary of AMF PDM results and methods , Results tab, cell H55.
- A 36-month follow-up study of non-AMF nets conducted by the National Malaria Control Program and the Kinshasa School of Public Health in DRC found that "overall survival in serviceable condition during the last survey [at 36 months] was 37% in Sud Ubangi and just 17% in Mongala." USAID, Durability monitoring of LLINs in Democratic Republic of Congo, Final Report, after 36 months follow-up, July 2019 (archive), Pg 4.
AMF hopes to move up the distribution to the second half of 2021 to get closer to its ideal 30-month distribution cadence. Replacing the nets after 30 months is only slightly less cost-effective than replacing the nets after 36 to 38 months because the nets do not provide much protection after 30 months of use.
- Guinea ($12.4 million). AMF currently estimates the size of the gap at $12.4 million. The distribution is scheduled for late 2021.
Sources: GiveWell's room for more funding analysis for AMF [June 2020], spending opportunities, AMF view; Against Malaria Foundation, conversation with GiveWell, July 2, 2020 (unpublished); Against Malaria Foundation, responses to GiveWell's questions, August 14, 2020 (unpublished).
 "AMF says the ideal lead time for its work is 23 months." The GiveWell blog, "Why we’re excited to fund charities’ work a few years in the future," August 25, 2020.
 We spoke with AMF and reviewed reports that it provided about its general monitoring methodology. We also spoke with the organizations that AMF partnered with to conduct monitoring in each of six countries (listed below), in order to understand how this methodology was implemented in practice. We focused our investigation on the methodology for post-distribution monitoring surveys (PDMs) for large-scale distributions from 2017 onward from which results are available. We believe these surveys are likely to be indicative of what we can expect from future distributions. Finally, we reviewed results from PDMs in the following countries (the PDM results we reviewed are in a private database and are not on the pages linked below, which list the dates and locations of the distributions):
- Against Malaria Foundation: distribution page, Guinea 2019 (archive)
- Against Malaria Foundation: distribution page, Ghana 2018 (archive)
- Against Malaria Foundation: distribution page, Malawi 2018 (archive)
- Against Malaria Foundation: distribution page, Togo 2017 (archive)
- Against Malaria Foundation: distribution page, Uganda 2017 (archive)
- Against Malaria Foundation: distribution page, Zambia 2017 (archive)
We plan to share our findings in more detail in the future.
 This grant was fully funded by a single donor who provided flexible funding for our top charities. The donor requested that we grant the funds immediately, rather than holding them to disburse with "Grants to recommended charities at GiveWell's discretion." We selected Malaria Consortium for the high cost-effectiveness of its planned work.
Taking the $8 million into account, we saw Malaria Consortium's remaining funding needs as less time-sensitive than AMF's when we considered granting the other flexible funds we received in the first half of 2020.
- Malaria Consortium's seasonal malaria chemoprevention (SMC) program. We estimate that Malaria Consortium's SMC program and AMF's net distributions are similarly cost-effective. GiveWell: 2020 cost-effectiveness analysis — version 1, "Results" tab, cells B20 and B26.
Malaria Consortium is stronger than AMF on the qualitative measures we use to assess charities. However, after examining Malaria Consortium's current plans and budget, we concluded that it doesn't have an urgent funding need at this time. Its next funding need is for work in 2022, and it doesn't need to secure funding for that work until late 2020 or early 2021: "Because of the lead time for procurement of SPAQ [medicine for seasonal malaria chemoprevention] we need planning security well in advance of the campaign, though I would point out that this relates to the time when discussions about the following year are happening with malaria programmes and other funders, rather than the time we place the order. Those discussions will start soon after the end of this year’s campaign, i.e. late this year/very early next year." Source: Email from Christian Rassi, SMC program director, Malaria Consortium, July 23, 2020, unpublished.
The majority of funding we direct is donated at the end of the year, so we will have another opportunity to meet Malaria Consortium's need in late 2020.
- HKI's vitamin A supplementation (VAS) program. HKI told us in June that it expected to need an additional $1 million per year for VAS in Kenya starting in the second half of 2020. According to HKI, the need emerged due to a reallocation of another funder's support from vitamin A to COVID-19 response. We considered using some of the $15.3 million to address this need.
We learned that it was likely that the other funder would in fact allocate funding to this program rather than letting it lapse (source: HKI conversation with GiveWell, August 13, 2020, unpublished). We may still want to provide funding to support HKI's program in the future, but we do not currently believe that this need is time-sensitive. As with Malaria Consortium, we will have another opportunity to meet this need during the end-of-year giving season. Source: Helen Keller International: Concept Note for Filling VAS Funding Gap in Kenya, unpublished.