Below, we share a breakdown of this calculation (see the chart in the "Money moved by charity" section of the blog post for details about the amount of money directed to each organization):
We directed $98.1 million to the Against Malaria Foundation (AMF), Malaria Consortium’s seasonal malaria chemoprevention program, and Helen Keller International's vitamin A supplementation program. We estimate that these life-saving charities each avert a death for between $3,000 and $5,000. See 2020 GiveWell cost-effectiveness analysis — version 2, "AMF," "Malaria Consortium," and "Helen Keller International" sheets, "Cost per life saved, after downside adjustments, adjustments for excluded effects, and funging adjustments" rows, available here for AMF, here for Malaria Consortium, and here for Helen Keller International.
Please note that the range we use here does not correspond perfectly with our cost-effectiveness analysis; we think it is a reasonable simplification. As we state on this webpage, "We present cost-effectiveness estimates [for our life-saving charities] as a range rounded to the nearest thousand dollars on our Top Charities page. This reflects the degree of precision we believe our model can estimate, as well as the range of cost-effectiveness that charities are likely to achieve across the countries they work in. Charities' cost-effectiveness can vary widely by geography, depending on the underlying burden of disease and the costs of operating in a given country."
We estimate 19,000 to 32,000 deaths averted as follows: $98.1 million divided by $5,000 = 19,620; $98.1 million divided by $3,000 = 32,700. We then round down each of the numbers to the nearest thousand.
We directed $11.5 million to the four charities we recommend for programs that support treatments for parasitic worm infections (deworming programs). Our cost-effectiveness estimates for these charities range from $0.72 to $1.13 per child treated. See 2020 GiveWell cost-effectiveness analysis — version 2, "Deworm the World," "END Fund," "SCI Foundation," and "Sightsavers" sheets, "Cost per person dewormed (per year)" rows, available here for Deworm the World, here for END Fund, here for SCI Foundation, and here for Sightsavers.
We estimate more than 12 million worm treatments by dividing the amount of the donations directed to each of the deworming charities by the cost per person dewormed (per year) for that organization, then adding the totals together.
- We directed $17.4 million to GiveDirectly for its cash transfer program. We estimate that 83% of these funds are ultimately transferred to households. The total transfer size per household is $1,000. See 2020 GiveWell cost-effectiveness analysis — version 2, "GiveDirectly" sheet, cells B5 and B7. We estimate the number of cash transfers by multiplying $17.4 million by 0.83 and then dividing by $1,000, resulting in a total of 14,442, then rounding down to the nearest thousand.
 Please note:
- We report on “metrics years” that run from February through January; for example, our 2019 data cover February 1, 2019 through January 31, 2020.
- In an effort to present a more comprehensive measure of our influence on charitable giving, we included GiveWell Incubation Grants in our headline "money moved" figure for our 2018 metrics report and continue to do so for our 2019 report. In previous reports, we excluded Incubation Grants from this figure.
 Our "headline money moved" figure only includes donations we are confident were the result of our research. For a list of the types of donations we count in our headline money moved, see this page.
 Our "best guess of total money directed to charities" includes two additional categories of donations about which we have more uncertainty than our headline figure:
- We estimated $15-25 million in additional donations that were not directly attributed to GiveWell but which we guess were due to our recommendations. We count $18.1 million in our "best guess of total money directed to charities" figure, but believe other amounts within this range would be reasonable.
- We estimated $4.1 million in donations where we believe our research played an important role. We expect that some but not all of this funding can be attributed to GiveWell and therefore count only 50% of this in our "best guess of total money directed to charities" figure.
For details on how we calculated these figures, see our full 2019 metrics report.
 In 2018, we tried to estimate how donations not attributed to individual donors would break down across the different donor size categories we track. However, we do not have strong evidence for the assumptions involved in this calculation and believe that it does not significantly improve our analysis. In 2019, we decided to only report money moved by donor size for donations that we can confidently attribute to individual donors.