Americans spent about $15.9 billion on the 2024 federal elections. The evidence says most of it changed almost no minds. Here is how a political donation, a charitable donation and a ballot compare on impact per unit of effort.
Kenya's 200-village basic income study found more income, more savings and no drop in work. Lump sums beat monthly payments. Cash is the benchmark everything else is measured against, and that is its problem.
A 2022 meta-analysis found water treatment cut under-5 deaths by about a quarter. GiveWell's own read is roughly half that, and it still funds the program at scale. Why the gap, and why both numbers are good news.
Carbon offsets often deliver little. Policy advocacy groups are estimated at under a dollar per ton avoided, in expectation. Here is the case, the caveats, and where the money went in 2025.
Group therapy programs in Uganda and Zimbabwe score highest on a new happiness-based metric. The estimates are real, contested, and worth understanding before you give.
Hundreds of millions of children have elevated blood lead. Getting lead out of paint costs a few dollars per healthy year of life, by one charity's own model. Here is what the evidence does and does not show.
GiveWell's four top charities save a life for an estimated $3,500 to $5,500. Here is where the figure comes from, what it leaves out, and why a claim of $10 or $50 per life should make you suspicious.
A capsule costs almost nothing. Getting it to every child under five, twice a year, and the fact that most of them would have lived anyway, is why the honest cost per life saved is in the thousands.
School deworming is nearly free and reaches hundreds of millions of children a year. The health evidence is weak. The long-run income evidence is strong but rests on one study. How to hold both.